Via Electronic Transmission
February 12, 2016
Wolf Plan Comments California Department of Fish and Wildlife [email protected]
To the Department:
1
The following comments regarding the “Draft Conservation Plan for Gray Wolves in California” (“Draft Wolf Plan” or “Plan”) are submitted on behalf of the following organizations and our combined total of more than 2.9 million California members and supporters:
Animal Legal Defense Fund California Wolf Center Cascadia Wildlands Center for Biological Diversity Defenders of Wildlife Endangered Species Coalition Environmental Protection Information Center (EPIC) Friends of the Wisconsin Wolf and Wildlife Howling for Wolves Humane Society of the United States Klamath Siskiyou Wildlands Center Living with Wolves National Wolfwatcher Coalition Natural Resources Defense Council Predator Defense Project Coyote Sierra Club California WildEarth Guardians Western Watersheds Project
INTRODUCTION
We recognize the extraordinary endeavor by many contributors that has resulted in the Draft Wolf Plan. We thank the California Department of Fish and Wildlife (“Department” or “CDFW”) for assembling a Stakeholder Working Group (“SWG”), for the many SWG meetings held by the Department over a several-year process to discuss key issues with SWG members and obtain their input, for the writing of this Draft Wolf Plan for public review, and for providing public comment opportunity via written comments and public meetings. We especially appreciate the opportunity you made available for the public to provide input on the plan by hosting meetings throughout the state, from Yreka to Long Beach.
There are many parts of the Draft Wolf Plan with which we fully agree and support. Other parts are not based on science; we do not support them and believe it would be at best a terrible mistake and at worst a travesty should those provisions be adopted in a final version of the Plan.
We are pleased that the Draft Wolf Plan covers a wide range of topics which are both critical for wolf conservation and essential for public knowledge and understanding as we welcome wolves back to the Golden State. We agree with the Draft Wolf Plan’s emphasis on nonlethal coexistence measures to deter or reduce livestock-wolf conflicts, which are more effective over the long term and far less costly than killing wolves or other predators.1
1 McManus et al., 2014; Imbert et al., 2016.
2
We also appreciate that the Draft Wolf Plan does not place a cap on the wolf population nor create wolf-and-no-wolf zones.
The Draft Wolf Plan’s chapter on disease is extremely informative and will, we hope, bring a halt to the baseless claims that wolves will ravage our state with disease. Nothing could be farther from the truth and this chapter does an excellent, scientific and easy-to-understand job of dispelling such claims and presenting the facts.
However, we are concerned that some key topics have been entirely left out and that nearly all of the published literature provided to the Department a year ago by the environmental conservation caucus of the SWG is neither discussed nor cited to in the Draft Wolf Plan. Additionally, key concepts and documents that were drafted, shared and edited by all interested members of the SWG are missing. Additional significant concerns expressed during the SWG process remain among the environmental conservation caucus groups and among the additional groups who have participated in crafting this comment letter.
In the following pages, we address the topics in this bulleted list:
1. CDFW has a legal duty under the Public Trust Doctrine to manage wildlife on behalf of all citizens of California.
2. The Plan should seek to recover wolves, not simply conserve and manage them.
3. The Plan’s tone should reflect that wolf recovery is a conservation opportunity, not a challenge to be overcome.
4. Promoting coexistence between wolves and livestock producers is of critical importance.
5. The Plan should prohibit the killing of wolves for depredations on public lands, require use of nonlethal measures before resorting to any lethal control of wolves, and must codify enforceable lethal take provisions.
6. The Plan should explain the correct use of livestock guarding dogs.
7. Depredation Investigations Protocols should be clearly articulated and included in the Plan.
8. Thresholds for population numbers and duration of time for phase transition are inconsistent, too low and not scientifically justified.
9. The threshold for seeking state-delisting is far too low and not scientifically defensible.
10. Seeking federal down-listing in protection levels and/or state legislative permission to obtain kill authority potentially creates confusing conflict between federal and state law, sets dangerous precedent and is unwarranted.
3
11. Outreach and education efforts should include compliance-enforcement information.
12. The Plan needs to prioritize recovery, conservation and management actions and prioritize securing funds from state and federal sources for implementation.
13. The Plan must include a comprehensive plan of action for public education aimed at recipients of wolf-location information.
14. The Draft Plan lacks key information referenced in the Draft Plan and/or which was discussed and intended by SWG members to be included in the Plan.
15. Wolves, coyotes and bears should not be killed to conserve wild ungulate populations.
16. Threats to wolves from illegal killing due to mistaken identification as coyotes have not been adequately addressed.
17. The Plan includes no discussion of potential economic benefit to local and regional economies from reestablishment of wolves, wolf-related ecotourism and consumer market for predator-friendly raised livestock products.
18. Comments regarding the Plan’s assessment of wolf taxonomy, population size and genetics issues regarding hybridization.
19. Ungulate population and habitat management are important aspects of wolf conservation and recovery efforts.
20. CDFW should actively seek out all opportunities to weigh in on land management actions with federal agencies and participate in land management planning processes.
21. The Plan should describe priorities for protecting, restoring and enhancing habitat that would benefit wolves because the State Wildlife Action Plan identifies the gray wolf as a Focal Species of Conservation Strategies.
22. The Plan should identify habitat conservation and connectivity priorities that will benefit wolf recovery.
23. The Plan’s trophic cascades discussion should include published research demonstrating wolves’ positive impacts in the Western Great Lakes states.
24. The Plan’s discussion on impacts of wolf mortality and wolf-killing on wolf packs should include the findings of a 2014 symposium on this very topic.
25. The Plan’s discussion of human social tolerance for wolves should address and cite to additional sources.
4
26. The Plan’s discussion of human perceptions and interactions with wolves should include discussion and citation to new paper establishing that the majority of attacks on humans by carnivores is due to inappropriate conduct by humans.
27. Evidence of historical wolf presence in California as indicated in languages, tales, practices and ceremonies of Native Peoples deserves a heading other than “Anecdotal Observations.”
COMMENTS
CDFW Has a Legal Duty Under the Public Trust Doctrine to Manage Wildlife on Behalf of All Citizens of California
The State of California has a legal duty to manage its natural resources, including wildlife, in a manner that benefits all of its citizens. This duty is derived from California’s statutes and a long common law tradition requiring each state to protect and preserve the natural resources shared by its citizens called the public trust doctrine.
Common law principles reaching back to antiquity place a duty on the state, as part of its sovereign nature as the representative of the people, to hold common natural resources in trust for its citizens.2 This trust requires the state to preserve natural resources, and to protect its citizens’ interests in those resources, by safeguarding against their exploitation for private gain at the expense of the public good.3 Historically, the public trust doctrine arose to protect the public’s right to access tidelands and navigable waters, specifically for their use in navigation, commerce, and fishing.4 Over time however, the public trust duty has expanded beyond its traditional boundaries. In California, the public trust duty of the state includes the protection of wildlife resources.5 California courts have reached this conclusion directly, citing the important shared resource provided by wildlife.6 California Courts have also reached this conclusion implicitly through the recognition that the prudent allocation of other natural resources–namely State waters–requires the State to consider the effect of its decision making on wildlife.7 In addition, California Fish & Game Code explicitly states that wildlife resources are held in trust by the State for the benefit of its citizens.8 As such, it is clear that California law treats wildlife as an important natural resource that provides significant public benefits and therefore necessitates State protection through a public trust.
2 National Audubon Society v. Superior Court of Alpine County, 33 Cal.3d 419, 433 (1983). 3 Berkeley v. Superior Court, 26 Cal. 3d 515, 521 (Cal. 1980); See Illinois Central Railroad Company v. Illinois 146 U.S. 387 (1892). 4 See Illinois Central Railroad Company v. Illinois 146 U.S. 387 (1892) (In the past, the public trust doctrine limited the state’s power to alienate submerged land and acted as a safeguard against the exploitation of those resources for private gain precluding public access). 5 See Center for Biological Diversity v. FPL Group, Inc., 166 Cal. App. 4th 1349 (2008). 6 Id. 7 See National Audubon Society, 33 Cal.3d 419, 433 (1983). 8 California Fish & Game Code § 711.7(a); California Fish & Game Code § 1600.
5
Because the State represents its citizens in its sovereign capacity, CDFW must exercise its control over wildlife pursuant to the public trust for the benefit of the people as a whole, not only for the benefit of livestock owners, hunters, or individual landowners. The Draft Wolf Plan fails to consider the intrinsic value of wolves as a part of wildlife under the Public Trust Doctrine, thereby necessitating their protection and not simply their management. The Draft Wolf Plan examines the potential negative impacts of wolves on the environment without considering the potential benefits of wolves on the ecosystem. Under the Public Trust Doctrine California’s citizens have the right to the aesthetic enjoyment of wildlife and ecological benefit that strong predator populations provide.
California Fish & Game Code § 1801 declares that it is the policy of the state to encourage the preservation, conservation, and maintenance of wildlife resources under the jurisdiction and influence of the State. This section states that fulfilling the objectives of this policy requires the perpetuation of wildlife for their intrinsic and ecological value as well as their more direct benefits to California residents. In contrast, the Draft Wolf Plan explicitly declines to preserve or conserve the impending wolf population, thereby violating CDFW’s obligations under Section 1801.
California’s Wolf Plan Should be a Recovery Plan, Not Simply a Conservation and Management Plan
The Plan provides for conserving, information-gathering and managing wolves in a 3-Phased approach but does not provide for active “recovery” efforts, despite the fact that in the midst of the SWG process the gray wolf was listed as endangered under the California Endangered Species Act (CESA). The environmental caucus repeatedly raised this issue during the SWG process, to no avail.
The Draft Wolf Plan asserts that CESA does not provide for preparation of recovery strategies except for one aquatic species. (Part I, p. 10.) However, CESA states as follows:
“[I]t is the policy of the state to conserve, protect, restore, and enhance any endangered species or any threatened species and its habitat . . . .” (F&G Code section 2052) The inclusion of the phrase “restore and enhance” is not mere surplusage but instead informs that it is state policy to take actions for listed species beyond conserving and protecting them. It is an implicit mandate for recovery of the species.
At the public meeting held in Long Beach on January 26, 2016, a Department representative told the public that there isn’t enough information available about wolves in California to define “recovery” of wolves. If this is the case, it is all the more troubling that the Draft Wolf Plan provides for consideration of delisting the species when the wolf population reaches 50-75 animals. (Part I, p. 21.) Delisting implies that CESA’s protections are no longer needed, i.e., that the species is recovered. By no measure would a wolf population of 50-75 animals be considered biologically recovered and the Department cannot have it both ways. Either there is not yet sufficient scientific information about wolves in California to know what recovery would be and
6 therefore no population threshold for recovery can yet be set, or there is ample scientific information about wolves in California to propose a threshold number.9
All evidence points to the first option. Much of that evidence comes from the Department’s own statements which appear repeatedly throughout the Draft Wolf Plan indicating that evidence of historical wolf distribution and abundance is speculative, and that California’s landscapes, wild ungulate population numbers and human density are so vastly different from other states which have wolves that information from those states regarding wolves cannot be relied upon to be accurate predictors of how things will play out for and with wolves in California.
Actions and strategies proposed in the Draft Wolf Plan are aimed at conservation of an established wolf population and management of an establishing -- and then established -- wolf population. We recommend the Department revisit all of the action strategies set forth in the Plan and reassess what changes could be made that would aim instead for recovering the species.
The Return of Wolves to California is a Historic Conservation Milestone and Cause for Celebration and the Tone of the Plan Should Reflect This.
In California, as in almost every state of the coterminous United States, the gray wolf was driven to extinction by the early 1900’s due to a concerted effort to eradicate the species on behalf of the livestock industry. The fact that the gray wolf is now returning to California is a remarkable event and a testament to the power of the federal Endangered Species Act to bring a species back from the brink when there is political willpower to do so. When wolf OR-7 from Oregon lifted a paw on the Oregon side of the border and set it back down on the California side of the border, he made history and international headlines. Media headlines throughout California laid a welcome mat for this wolf and for the wolves that would follow. After a nearly 90-year absence, the gray wolf is returning to California and the state has an opportunity to right a historic wrong. The tone of the state’s wolf Plan should reflect that the return of wolves heralds a historic moment in conservation history in California and an incredible opportunity to restore a species whose presence and natural hunting practices lead to healthier, more biodiverse ecosystems.
Instead, the Draft Wolf Plan’s tone regarding wolves is dry and filled with worry and reservations. Its pages contain words like “challenge,” “challenging” and “concerned.”10 From all of our organizations’ combined reading of the 311-page document, we noted only one sentence which expresses a view from the drafters of the Plan that wolves might be a positive
9 We agree with the Department’s public statement that there is insufficient scientific information at this time specific to wolves in California to know what would constitute recovery of the species in California – and therefore no delisting threshold should be proposed at this time. We elaborate further on the Department’s proposed delisting threshold in a subsequent section of this comment letter. 10 The Draft Wolf Plan’s tone regarding elk contrasts starkly with how wolves are portrayed. In one of the opening paragraphs of Chapter 6, elk are described in glowing terms as one of California’s most important visible natural resources, a significant part of the food chain and a highly-valued species for viewing and hunting. Since elk became the majestic animals they are due to their coevolution with equally magnificent predators including their primary predator, the wolf, a few such similar sentences regarding wolves could be sprinkled throughout the Draft Wolf Plan.
7 addition to California. On page 17 of Part I, the Draft Wolf Plan posits: “Most interactions between wolves and the public will likely consist of memorable observations.” This magnificent, charismatic and ecologically-important species, the gray wolf, deserves much greater acknowledgment of its significance, beauty and majesty than a one-sentence homage. The Draft Wolf Plan’s perspective needs an attitude adjustment. We believe the vast majority of Californians who are aware that wolves are returning to our State agree with us.
The Plan should notify readers that scientists the world over are calling for the protection and recovery of apex predators like wolves and that the return of wolves is cause for celebration. Apex predators around the globe are in significant decline due to persecution by humans; their decline has serious detrimental effects on the planet’s biodiversity, which in turn impacts human health and well-being.11 The importance of top-level predators in their ability to help moderate impacts from climate change is even a subject of scientific agreement.12 The Plan should discuss the published literature on this topic and should frame the protection and recovery of wolves in California as a welcome and essential action for wolves and humans alike.
Promoting Coexistence between Wolves and Livestock Producers is of Critical Importance.
Promoting coexistence between livestock producers and wolves is of critical importance for successful wolf recovery in California. We are especially grateful that that the Draft Plan has a strong emphasis on the use of proactive measures for protecting both livestock and wolves.
The use of nonlethal management tools to reduce wolf-livestock conflicts is the key to successful coexistence between ranchers, rural communities and wolves, and to the success of CDFW efforts to manage wolves effectively for all constituents.
In order to have an effective nonlethal management effort, more than a description of the tools is needed. The ranching community will need help to both learn how to use the tools effectively and to properly implement their use on the ground. Success is more than just knowing and having the tools. Success will come through understanding, education, training, local on-the- ground assistance, and local and state level support.
It is impossible to list all the individual actions, education tools and printed materials, as well as CDFW and outside support needed to make a nonlethal program successful. However, the nongovernmental organization (“NGO”) community in consultation with the ranching community is working to produce a more complete program that CDFW should use as a foundation for developing a state sanctioned nonlethal management and conflict reduction program. Several NGOs have shared with CDFW the framework for this program, entitled the “California Wolf-Livestock Risk Management Plan Framework.” This was shared with the Department in a meeting with Karen Kovacs and Eric Loft in November 2015.
11 Estes et al. 2014; Ripple et al. 2014. 12 Urban and Deegan, 2016. http://mobile.nytimes.com/2016/02/06/opinion/t-shirt-weather-in-the- arctic.html?_r=0; Wilmers and Getz, 2005.
8
This document is a work in progress at this time. However, the NGO community is continuing to develop this program and make it very specific and useful for CDFW and ranchers. A subset of the signatories of this letter intend to have the program materials compiled by late spring of this year. We ask CDFW to work with interested NGOs to continue to develop and refine this program to help make it a something that will work for livestock producers and the Department, and meaningfully contribute to reduced conflicts between wolves and livestock. Once the program framework is more complete, we recommend the state review, edit, and develop it so it can be formally incorporated into the Wolf Conservation Plan.
The information in this conflict reduction program will need to be available, as well as a process to support livestock producers to implement the program. In the plan will be some suggestions on funding sources, from both government and private entities.
It is critically important that CDFW supports this process, and has plans to implement a thorough and well thought out conflict reduction plan. Everyone wins when wolf livestock conflicts are minimized.
California Must Not Kill Wolves for Depredations on Public Lands, Must Require and Rigorously Use Nonlethal Coexistence Measures Before Resorting to Lethal Control of Wolves, and Must Codify Enforceable Lethal Take Provisions.
If there is chronic depredation and correct use of nonlethal measures, and if that depredation is occurring while the livestock are on public land, it is not appropriate to kill those wolves whether the wolves, themselves, at the time of the Department’s consideration of removal, are on public or private land. Those depredations, and the nonlethal measures taken to prevent them, should be costs of doing business on public land. As noted below, this is expressly one of the reasons that grazing fees are set so low. The key question is where the livestock were at the time of depredation.
Lethal control of wolves for chronic depredation of livestock should be a last resort and taken only after all reasonable efforts have been exhausted to correctly employ feasible nonlethal methods, strategies and tools, and only in the case of chronic depredations (i.e., multiple depredations by the same pack or individuals). Further, lethal control of wolves in response to depredations on public lands is not acceptable. Public lands are owned by all members of the public, and public lands and wildlife are held in trust by state and federal agencies for all members of the public.13 Public lands grazing rates have been set at a rate nearly 20 times lower than the cost to rent private lands for grazing, and American taxpayers subsidize the use of those lands at a cost of more than $120 million annually.14 One basis for setting fees so low, as noted
13 As explained more thoroughly in the last section of this letter, the Public Trust Doctrine requires that the State manage its natural resources, including public lands and wildlife, to the benefit of all of the people of the State. 14 Glaser, C., Romaniello, C. and K. Moskowitz. 2015. Costs and Consequences: The real price of livestock grazing on America’s public lands. Special Report. Center for Biological Diversity. http://www.biologicaldiversity.org/programs/public_lands/grazing/pdfs/CostsAndConsequences_01- 2015.pdf
9 in a 1977 report jointly issued by the Secretary of the Department of the Interior and the Secretary of Agriculture, is to account for the fact that on public lands there may be some losses of livestock due to predators.15 Public lands are also frequently the very habitat where wolves can find their chief wild ungulate prey species, deer and elk. The killing of wolves and other native predators for livestock losses or for preying on elk or deer on public lands is unacceptable. Any actions or strategies the Department includes in the final Plan or implements on the ground must not include the killing of wolves (whether on private or public land) in response to depredations that occurred on public lands.
We understand from the Draft Wolf Plan that the Department intends to use lethal control of wolves in cases of chronic livestock depredation, after first employing nonlethal methods, tools and strategies, if the Department has lawful authority to do so. There are some very important lessons to be learned for California from wolf management examples set in neighboring Oregon and Washington. The most important of these is the codification of the wolf plan as an administrative rule, and for future provisions concerning lethal take of wolves, that such provisions also are codified.
In many respects, Oregon and Washington have similar wolf plans. However, Oregon has both statutes and agency rules governing wolf conservation, whereas Washington’s plan is merely a non-enforceable guidance document.
The California Plan, when finalized, will reflect the agreements collectively arrived at by a diverse group of stakeholders. The Plan itself proclaims that it “covers key issues and potential actions CDFW believes important to the understanding and future conservation of wolves.”
California could avoid the mishandling of wolf conservation, and learn from previous mishaps by codifying the Plan’s provisions. Though both states have plans similar in substance, Oregon has seen substantial advances in wolf recovery with minimal conflicts, while Washington’s wolf management has been plagued by controversy.
Oregon has been the model state for wolf recovery. The Oregon Department of Fish and Wildlife (ODFW) has developed and codified in rule predictable and reliable responses to conflicts and various situation that arise with wolves. While Oregon has permitted killing of wolves in response to livestock depredations, such actions are governed by enforceable rules that leave no party involved guessing as to the response. This predictable arrangement also minimizes the political push and shove that inevitably occurs when there are conflicts or difficult situations.
For example in Oregon, by statute, livestock depredation is only “chronic” if the appropriate authorities confirm at least four qualifying incidents within a consecutive six-month period during Phase I (0-4 breeding pairs).16 Agency regulations require the livestock producer to prove that he or she removed “unnatural attractants of potential wolf-livestock conflict at least one
15Study of Fees for Grazing Livestock on Federal Lands. A Report from the Secretary of the Interior and the Secretary of Agriculture. October 21, 1977. 16 ORS § 498.014(a)(A)
10 week before the incident,” and that prior to and on the day of the incident, he or she implemented at least one nonlethal measure deemed appropriate by ODFW.17
These rules governing lethal take of wolves in response to depredation has led to minimal political squabbling, an increasing wolf population (the state has not spent tax payer dollars killing wolves since 2011) and, due to the incentives for increasing preventative measures, depredations have decreased.
The Washington Department of Fish and Wildlife (WDFW) also spent considerable time developing a wolf plan, a document that incorporated the views of a 17-member stakeholder group, 65,000 written comments, and 23 scoping meetings. However, WDFW failed to codify the provisions in rule, and following the Washington Fish and Wildlife Commission’s adoption of the Plan in 2011, the Commission decided to deny a petition for rulemaking to codify lethal control provisions of the Plan. The Commission reasoned that (1) determining the need to use lethal control to stop repeated depredations is a complicated issue, and (2) limiting the flexibility articulated in the Plan reduces the ability to address each case-specific conflict.
However, this flexibility has caused considerable problems for the state, and the discretion so desired by WDFW has led to massive conflict, state legislative investigations, numerous legislative battles, and public controversy. When problems arise with implementing the wolf plan, inevitably there are going to be forceful voices on all sides lobbying for a certain outcome. When there are dead animals involved, these voices tend to get very loud. Having provisions of the wolf plan codified in rule allow a state agency to stick to the Plan and gives state employees a defensible plan of action. Too much discretion can lead to bad outcomes.
As examples, two nearly identical lethal control mishaps in Washington illustrate the need for legally enforceable lethal take provisions. In 2012, WDFW exterminated the 7-member Wedge Pack, costing taxpayers $76,500. Despite the legislative mandate to “preserve, protect, [and] perpetuate” wolves as “[w]ildlife . . . property of the state,”18 the commission elected instead to exercise its discretion to benefit the economic interests of a single individual. That individual was grazing his cattle on publicly-owned national forest land, without taking the appropriate nonlethal measures to protect his herd. The killing of the pack led to massive public outcry, administrative rule-making petitions, and a legislative investigation. A spokeswoman for Phil Anderson, then-director of WDFW, said he “never wants to do this again.”19 Clearly, WDFW’s decision making would have benefitted greatly from clear standards governing agency response.
One month later, WDFW once again chose to subvert the Plan’s standards for lethal take. The Huckleberry Pack situation was eerily similar to the circumstances surrounding the Wedge Pack. The rugged terrain leased by the rancher from a private timber company was ill-suited to sheep grazing. The rancher had experienced no depredations prior to late June 2014. A herder who had been managing the flock quit that summer at some point before the depredations occurred. This
17 OAR 635-110-0010(8)(b)(B) 18 RCW § 77.04.012 19 Cassandra Profita, The Cost of Killing Washington’s Wedge Wolves, Oregon Public Broadcasting, Nov. 14, 2012, http://www.opb.org/news/blog/ecotrope/the-cost-of-killing-washingtons-wedge-wolves/
11 same rancher declined nonlethal conflict avoidance resources offered by WDFW and Washington State University earlier that spring. Eventually some depredations were discovered.
Although it was still unclear whether wolves were responsible and though it was likely the rancher’s utter disregard for the plethora of resources offered to him that led to the depredation, WDFW agreed to kill four pups through aerial gunning based on the thought that less mouths to feed would result in less depredation. Unfortunately, the USDA/Wildlife Services sharpshooter hired by WDFW to carry out the kill order mistakenly killed the breeding female. The department embarked on its aerial gunning operation in the early morning hours of the weekend without notice to the public and, when contacted, officials at the department indicated they would not respond to comments or questions until the following week. The Huckleberry Pack fiasco occurred under Phil Anderson, the same director overseeing the department at the time of the Wedge Pack disaster. Again, legislative inquiries were launched, there was massive public outcry, and numerous bills lined up for the upcoming legislative session aimed at targeting the agency’s funding.
These are clear examples of how discretion regarding wolf conservation and management can lead to horrible decisions and ongoing conflict with potentially devastating implications for a wildlife agency. The California Fish and Game Commission should learn from the examples set by Oregon and Washington. Codifying plan provisions sends a clear message that the will of the people of the state of California, embodied in the Plan, shall determine the department’s course of action for wolves. Enforceable provisions provide the department with a shield to defend itself against various interests when attempting to develop plans of action in difficult situations that will inevitably occur. Establishing enforceable boundaries compels discourse and collaboration between parties holding opposite views with respect to wolves.
During the SWG process, the environmental caucus presented the Department and fellow SWG members with proposed regulatory language for codifying the lethal take provisions of the Wolf Plan. Tables in Appendix G refer to an “Operational framework for lethal control” and provide some descriptions of Options/Actions but nowhere does the Plan propose any specific, legally enforceable regulatory language on the use of lethal control of wolves for chronic depredation of livestock. We have included in Appendix A of this comment letter, the proposed regulatory framework prepared by the environmental caucus.
The Plan Should Explain the Correct Use Of Livestock Guarding Dogs.
Part II of the Draft Plan at p. 122 includes a section entitled Predicting the Potential Effects of Wolves on Livestock and Herding/Guard Dogs in California. We present a different perspective which we think is more accurate.
Livestock Guardian Dogs (LGDs) are most effective as sentinels for sheep and cattle when the livestock are bunched up or within fencing during the day or night. There are many existing breeds currently available in the U.S. that have proved effective at alerting humans about predator presence. Examples are Great Pyrenees, Spanish Mastiffs, Pyrenean Mastiffs, Maremma, Anatolian Shepherd, Akbash and others. It is not important, or desirable, to have
12
extremely aggressive fighting dogs as LGDs. No dogs should be expected or encouraged to fight wolves. Rather, the LGDs serve to alert humans, on site, about the presence of predators. For this reason it is important to have dogs that have instincts and training to stay with the flock or herd, in sufficiently large numbers on site to act as sentinels and deterrents.
It is thought that wolves moving through an area will avoid livestock surrounded by a sufficient size “pack” of LGDs. The LGDs should be trained to stay with the pack rather than roam across the open terrain. Single LGDs on the open range are not expected to serve a useful purpose. LGDs in combination with other tools serve to discourage wolves from seeing livestock as prey. LGDs should be thought of as deterrents rather than “protection” against wolves especially when used in combination with tools such as fencing, fladry, removal of boneyard attractants, animal husbandry techniques and lighting such as Foxlights. LGDs in combination with human presence are an effective tool to avoid negative interactions between wolves and livestock in the appropriate setting.
In summary, the Plan seems to regard LGDs as fighting protectors of livestock. We do not believe this is the appropriate way to view them.
Depredation Investigations Protocols should be Clearly Articulated and Included in the Plan.
As noted in the Draft Wolf Plan, wolf depredations on livestock in western states comprise a small fraction of all livestock losses. However, when a wolf-caused depredation is suspected and reported to officials, the ensuing investigation by agency staff is a matter of concern to all the public. The outcome of the investigation is important to livestock producers, conservationists and the general public and, because the outcome could end up designated as a strike against a particular wolf or wolves it is important to the lives of wolves, as well. Thus it is essential that investigations not be left to the whims of whoever is in charge in a particular circumstance.
It is essential that in the Plan the Department enumerate defensible procedures for training of investigators, the investigation itself, and criteria for determinations. The depredation investigation protocol should also set forth requirements and procedures for documentation and types of documentation, and for transparency to the public. It may also be necessary to enumerate procedures for review; if so, any third-party reviewer needs to be qualified and unbiased. The local vet, the local sheriff, the local USDA/Wildlife Services agent are not.
Some people want to see every dead animal blamed on wolves; others, none. What is most important is that California gets it right, that California’s Wolf Plan includes a definitive protocol, and that the decisions be defensible and transparent so that conclusions can be verifiable.
13
Thresholds for Population Numbers and Duration of Time for Phase Transition are Inconsistent, Too Low and Not Scientifically Justified.
As a preliminary matter, we note that the number of breeding pairs (“BP”) specified to mark phase shifts in the Draft Wolf Plan’s adaptive management strategy are inconsistent throughout the document and thus confusing to the reader. Specifically we note the following inconsistencies which need to be rectified:
Part I of the Draft Wolf Plan, at p. 21 states that: Phase 1 ends at 4 BP Phase 2 starts at 5 BP Phase 3 starts at 9 BP
Part II of the Draft Wolf Plan describes these thresholds in a different fashion, in two different places. Part II at p. 272 states that: Phase 1 ends at 4 BP Phase 2 starts at 4 BP Phase 3 starts at 6 BP
Part II at p. 283 states that: Phase 2 starts at 4 BP Phase 3 starts at 8 BP
At a substantive level, the thresholds the Department is proposing for numbers of BPs and transitions between management phases is not scientifically-based, have not been adequately justified by the Department, and are unacceptably low. The transition also is proposed to occur after an insufficient period of time has passed to best ensure reliable predictions of the population trend and that breeding pair numbers won’t immediately decline. The Department asserts throughout the Draft Wolf Plan that California’s landscape, prey base, and human population dynamics are different from other states where wolves are reestablishing. So let’s wait for the science the Department and other researchers develop regarding wolf reestablishment in California over time, before setting numbers goals for Phase shifts (and for delisting).
Threshold numbers should not be set at this time, but instead the Phase I period will allow for information to be gleaned over time until a shift in strategies is warranted. If a Phase I goal is set now, a precautionary approach should be applied, and the Phase I goal should not be less than 12 breeding pairs for at least three consecutive years to allow the Department time to gather sufficient data to determine if that’s even an adequate threshold. The shift from Phase II to Phase III should be left open until we know more about how wolves do in California as they populate the California landscape.
Additionally, time spans between shifts are of too short of duration to reliably indicate population trends and threats to wolf recovery and conservation. The Department should first conduct years of monitoring and data gathering to analyze trends in wolf and prey populations and distributions, among other factors. While Washington’s and Oregon’s state wolf plans set a
14 minimum duration of three consecutive years at a specified population level before shifting into a next phase of management strategies, the California Draft Wolf Plan inexplicably sets a time threshold of but two consecutive years. The Department provides no scientific justification for proposing a two-year period. Nor does it provide any scientific justification for lessening the time threshold from that employed by Oregon and Washington.
One thing we have learned from other states, as they have gone through or are now going through the early stages of wolf recovery, is that truly adopting proactive nonlethal coexistence methods, tools and strategies takes time. It especially takes time for livestock producers to embrace the concept of coexistence and accept it on a deeper, more cultural level, versus temporary willingness to accept the use of coexistence measures only because they aren't allowed to shoot wolves. Moving too quickly through the phases of an adaptive management approach to wolves can undo all the hard work to implement proactive methods and for nonlethal coexistence measures and philosophy to get a solid and accepted footing in the livestock community. It benefits no person and does not benefit wolves to rush through the phases because of political pressure or some preconceived notion that it is the best approach. Wolves and nonlethal coexistence measures must be given a real chance to succeed.
The Threshold of 50-75 Wolves to Consider State-Delisting is Far Too Low and Is Not Scientifically Defensible.
The Draft Wolf Plan’s proposal to consider state delisting at 50-75 wolves (Part I, p. 21) is not based on science. Instead it appears that the Department has settled on these numbers by drawing upon the state Wolf Plans for Oregon and Washington and then setting the bar even lower. The population goals and delisting thresholds in Oregon’s and Washington’s Wolf Plans have been found to be inadequate by most scientists who have evaluated them. There is no scientific rationale to justify thresholds for delisting in California that are even more deficient.
Washington’s wolf Plan divides the state into thirds, sets numerical breeding pair goals for each third of the state and a time duration for which those goals must be maintained before delisting can occur.20 A decision by the Washington Fish and Wildlife Commission to delist wolves must be made “solely on the basis of the biological status of the species being considered, based on the preponderance of scientific data available.”21 And, “ [a] species may be delisted only when populations are no longer in danger of failing, declining, are not longer vulnerable . . . or to meet recovery plan goals, and when it no longer meets the definition [of endangered].”22 Washington’s Plan requires the existence for three consecutive years of 15 successful breeding pairs. Of those 15 successful breeding pairs, there must be 4 successful breeding pairs in each third of the state plus an additional 3 successful breeding pairs anywhere else in the state. Thus Washington’s wolf Plan sets delisting numbers and time span durations which are significantly higher than what is proposed in the California Draft Wolf Plan and requires a distribution of the wolf population across the entire state. Washington’s Plan also provides an option for gradually
20 Wolf Conservation and Management Plan. Wiles, GJ., Allen, H.L. and G.E. Hayes. Washington Department of Fish and Wildlife Wildlife Program. State of Washington. December 2011. 21 WAC 232-12-297, §4.1 22 WAC 232-12-297, § 4.2
15 reducing protections over time from endangered to threatened to state-sensitive to delisted, a safety net mechanism which is not proposed at all in California’s Draft Wolf Plan.
Oregon’s Wolf Plan includes a phased management approach, divides the state into halves, and sets numerical breeding pair goals for each half of the state and a time duration for which those goals must be maintained.23 The “conservation population” objective in Phase I for each half of the state is defined as 4 successful breeding pairs for 3 consecutive years; the “management population” objective in Phase II for each half of the state is defined as 7 successful breeding pairs for 3 consecutive years; and in Phase III for each half of the state the Plan’s objective is to ensure the wolf population does not decline below Phase II levels. Because Oregon’s wolf population was reestablishing via dispersal westward from Idaho into eastern Oregon, when the state’s Wolf Plan was drafted, it was presumed that Phase I objectives would be met in the eastern half of the state prior to meeting separate Phase I objectives in the western half of the state (which could occur only once dispersing wolves made their way into western Oregon).
Oregon state law does not allow for delisting of a species in only a portion of the state. Thus Oregon’s Wolf Plan provides that consideration of delisting gray wolves can be undertaken when there are 4 successful breeding pairs for 3 consecutive years in the eastern half of the state, and that if state-delisting occurs at that point, the wolf population in the western half of the state will be managed by regulations as though the western half of the state were still fully state- endangered. Only upon reaching 4 successful breeding pairs for 3 consecutive years in the western half of the state may the wolf’s west-side population be managed as though no longer state-listed.
The Oregon Endangered Species Act requires that before a species may be delisted, the state Fish and Wildlife Commission must evaluate five enumerated delisting criteria and determine that none of them any longer present a threat to the continued existence of the species. The Act also requires that the Commission’s decision be based on documented and verifiable science. If the Commission is relying on data collected by and reports prepared by ODFW, for these to qualify as verifiable requires that the Commission engage an outside scientific peer review panel to evaluate those data and reports.24
In November 2015, ignoring the best available science and the law, the Oregon Fish and Wildlife Commission voted to state-delist wolves in Oregon.25 Three conservation groups have filed a legal challenge and the case is currently pending.26 At the time of the delisting vote by the Commission, ODFW estimated that Oregon’s wolf population stood at 82 confirmed observed individuals (ODFW reported 85 confirmed wolves as of mid-July 2015, but in the following two
23 Oregon Wolf Conservation and Management Plan. Oregon Department of Fish and Wildlife. December 2005 and Updated 2010. 24 ORS §§ 496.171 - 996 25 ODFW news release: Fish and Wildlife Commission delists wolves statewide in split vote (4-2). http://www.dfw.state.or.us/news/2015/November/110915.asp 26 Center for Biological Diversity press release: Legal Challenge Filed Over Removal of Protections From Oregon's Gray Wolves. http://www.biologicaldiversity.org/news/press_releases/2015/wolf-12-30- 2015.html
16 months one of these animals was illegally killed and two others were found dead under suspicious circumstances, reducing the known population to 82 animals as of the November delisting decision). Ninety percent of these confirmed wolves reside in eastern Oregon. In western Oregon there exists only one known breeding pair, the Rogue pack (wolf OR-7’s pack), and he and his mate have qualified as a successful breeding pair for only two years so far. It will be several more years before western Oregon has 4 successful breeding pairs, and several more years after that before achieving the duration benchmark of at least 4 successful breeding pairs for 3 consecutive years. By the time western Oregon’s wolf population may be managed as though no longer state-listed, the overall state wolf population will likely be double the number of wolves which existed at the time the Commission made its delisting vote. Unless, of course, the statewide delisting and transition to Phase II and then Phase III management strategies results in more killing of wolves by agency actions and by legal and illegal killing of wolves by private citizens, which result in an overall state wolf population decline and/or an inability of dispersing wolves to safely make it to the western half of the state.
Thus Oregon’s Wolf Plan sets delisting numbers and time span durations which are significantly higher than what is proposed in the California Draft Wolf Plan and requires a distribution of the wolf population across the entire state. It also includes regulatory mechanisms for continuing to protect and manage wolves as though still state-listed in the entire western one-half of the state to which wolves are just starting to make their way. California’s Draft Wolf Plan has no similar protective regulatory mechanism to keep dispersing wolves safely protected after an initial population of 50-75 wolves establishes. Since California’s wolves are arriving as dispersers from Oregon, an initial population of 50-75 wolves most likely will first reestablish in California’s northernmost counties. If delisting were to take place at that population level, the Draft Wolf Plan contains no regulatory mechanism like Oregon’s to protect and best ensure the safe establishment of wolves which disperse further south in the identified suitable wolf habitat in the central Sierra Nevada.
The pending Oregon wolf-delisting lawsuit was filed because the Commission violated the Oregon endangered species act when it voted to delist the gray wolf. The Act’s delisting criteria were not met, and the Commission did not seek an outside unbiased peer review of ODFW’s own status review of gray wolves, as is required by the Act. As part of the public comment period leading up to the Commission’s November meeting, 26 highly-credentialed scientists submitted comments on their own. The scientists who wrote comments are among the most experienced professionals in the U.S. and abroad in the field of wolf biology and ecology, mammalogy, population viability analysis and human-carnivore conflict social science. The scientists resoundingly denounced ODFW’s status review, population viability analysis and recommendation to delist as being fundamentally flawed, not justified by science, counter to science, ignoring the chief threat to wolf recovery and failing to demonstrate that delisting criteria had been met. A key criticism was that a population of only around 80-85 wolves, inhabiting only 12 ½ percent of identified current suitable wolf habitat in the state could by no measure be considered recovered and in fact this status of population and range distribution demonstrated that wolves are still very much endangered in Oregon. The Commission unfortunately chose to ignore the comment letters sent to them by outside expert scientists, and instead relied on ODFW’s status review and some short remarks prepared by four scientists who
17 were handpicked by ODFW shortly prior to the hearing and whose remarks were not made known nor available to the public until the delisting hearing was already underway.
The comprehensive comments submitted by the 26 outside expert scientists are relevant to California’s Draft Plan. The Department proposes to consider removing state protections for the gray wolf throughout California when the species’ population reaches a population threshold even lower than that set by Oregon and without consideration for how much of its suitable range gray wolves may or may not be occupying at that point in time. We have compiled these comment letters and provided them to you in Appendix B.
Seeking Federal Down-listing in Protection Levels and/or State Legislative Permission to Obtain Kill Authority Potentially Creates Confusing Conflict Between Federal and State Law, Sets Dangerous Precedent and is Unwarranted
Beginning on page 6, the Draft Wolf Plan details the legal status of wolves in California, highlighting the fact that wolves are currently protected under federal and state law. The Draft Wolf Plan clearly considers this protected status as a burden, as it “affects the state’s ability to manage the species with respect to any possible use of lethal take for management.” It is highly concerning that the Draft Wolf Plan provides that the CDFW will consider petitioning US Fish and Wildlife Services (USFWS) to down-list wolves to “threatened” in California when two breeding pairs are documented for two successive years, if wolves in California are still federally listed as endangered. The Draft Wolf Plan does not state any scientific or legal basis upon which such a request to the USFWS would be made, other than that a down-list would make the task of wolf management in the state of California easier for CDFW, whom, presumably, would seek special status for wolves in the state under ESA Section 4(d), granting the State kill authority.
CDFW offers no explanation for a down-listing request nor does the Draft Wolf Plan state how or why CDFW can show that the wolf population in California is significant and discrete from the Oregon and Washington populations, thus justifying any 4(d) special status. Moreover, considering the Plan’s own detailed discussion of how wolves have crossed borders between these states, it is not foreseeable how CDFW could even make such an argument. The Draft Wolf Plan should not be considering taking steps to override federal determination of the protected status of wolves at this point in time merely to allow CDFW more discretionary authority as to how to best manage wolf population. Such an approach degrades the importance of the ESA and listing decisions and creates dangerous precedent for any state to put its individual interests above the best interests of preserving wildlife on a national level.
Even if wolves are federally down-listed and CDFW successfully obtains kill authority, under Section 4(d) of the ESA, if wolves remain listed under CESA, CDFW will still lack kill authority. Thus it is clear that in such a situation, CDFW intends to seek state legislative authority to kill wolves once “Phase 2” population levels are reached, even if the CESA legal status of wolves remains listed as endangered.27 At numerous points in the Plan, CDFW discusses potentially working through the State’s legislative process in order to obtain kill authority for wolves, despite the existence of protections under CESA. Again, it is misguided for
27 See Plan, Part I, pp. 21-22, Part II, p. 272 and Appendix G.
18
CDFW to consider circumventing decisions made about listing status under CESA in order to make its job of wolf management easier. If and when the population of wolves in California reaches a level at which CESA listing may no longer be necessary, proper procedures should be implemented under Section 670.1, Title 14 of the California Code of Regulations to change the legal status of wolves in the State. To do otherwise disregards the significance of species being listed under CESA and thus meriting State protection.
Outreach and Education Efforts Should Include Compliance-Enforcement Information
Compliance-Enforcement Information. The “Outreach Goals” section to “Inform the public” is beneficial. Disseminating facts to dispel rumors and myths and correcting falsehoods are important to all aspects of “Interactions” listed in the “KEY ISSUES for WOLF CONSERVATION” section (Part I). Due to the palpable hostility to wolf recovery programs by a misguided or mis-informed minority, along with a number of reported and well-documented illegal killings of wolves in other states, it is incumbent upon the Department to take extra precautionary measures for wolf conservation.28 Two outreach focus areas that might reduce potential illegal activities should be considered for inclusion in the Plan—compliance and penalties for violation. This may be accomplished by expanding the “Inform the public” section.
Code and Regulation Compliance. Expanding Outreach Goals to cover compliance information related to Fish and Game codes, as well as Federal Endangered Species Act (ESA)29 and California ESA (CESA)30 regulations would be prudent and helpful to the public. The Draft Wolf Plan emphasizes that implementation of any of the strategies must always reflect the legal status of wolves, but the public, as well as livestock owners and sport/trophy hunters, may not be fully aware of more restrictive regulations with listed species protection. The Plan should inform all citizens of ESA’s and CESA’s legal obligations in the event of wolf (or any listed species) interaction. This type of educational outreach information is slightly different from implementation Strategy 7, which seems to focus on the public’s knowledge of wolves and attitudes—also important and worthwhile.
Enforcement—Potential Fines and Penalties. When enforcement reaches citation levels, the public should be apprised of the penalties for violations of ESA and CESA. This information should be included in the Plan:
ESA: Violations may be punished with fines up to $50,000 and/or one year imprisonment for crimes involving endangered species, and $25,000 and/or six months
28 In a “Mexican Wolf Conservation Assessment” of 2010, the FWS reported that the “illegal shooting of wolves is the single greatest source of wolf mortality in the reintroduced population.” US District Court, Arizona, Tucson Division, Wildearth Guardians and NMWA v US Dept of Justice, Case 4:13-cv-00392- DCB, 5/30/13, p4, item 11. 29 Federal ESA: To “take” means “to harass, harm, pursue, hunt, shoot, wound, kill, trap, capture, or collect, or to attempt to engage in any such conduct.” 16 U.S.C. §1532(19). 30 CESA: Prohibits the take, possession, purchase, or sale of endangered, threatened or candidate species. CA Fish and Game Code defines “take” as to, or attempt to, "hunt, pursue, catch, capture, or kill."
19 imprisonment for crimes involving threatened species. Misdemeanors or civil penalties are punishable by fines up to $25,000 for crimes involving endangered species and $12,000 for crimes involving threatened species. A maximum of $1,000 can be assessed for unintentional violations. Rewards of up to $2,500 are paid for information leading to convictions.31
CESA: Penalties may be imposed for violations of CESA. For taking or possession of a fully protected mammal, the base fine is $5,000; additional fees to the state, county, courts and surcharge can bring the total bail to $20,000.32 CalTIP is a confidential secret witness program that provides a number of options (toll free number, 24/7; a website; cell phone texting; or smartphone APP) for the public to report poachers, polluters, or any wildlife violation. If the information leads to an arrest, the caller becomes eligible for a reward. CalTIP rewards are funded by donations; no state funds are used.
By having legal obligations described more thoroughly as well as some semblance of the range of penalties and bail upon conviction, the implementation of the Plan, and especially “Strategy 2—Assess and address threats to wolf conservation,” are more likely to be successful. Strategy 2, c, “Minimize wolf mortality from accidental killing, and 2, d, “Minimize disturbance at active wolf den and rendezvous sites,” are examples of strategies that would benefit from expanded or more in-depth consequential information. Such material does not have to be either threatening or oppressive, but rather educational, which may be helpful to the public. It may also serve as an indicator of how serious the ESA/CESA listings are and how invested CDFW is in wolf conservation.
“Law enforcement” is mildly referenced throughout the Draft Wolf Plan in different roles (communication, presence to reduce poaching, enforcement of game laws, etc.). However, in order for the Plan to reach its goals, the law enforcement component should and will play a much greater role than may be implied. In fact, it may be the linchpin with regard to successful wolf conservation outcomes in light of the aforementioned wolf hostility. In addition to the need to increase wildlife officer staff for law enforcement, the Plan should confirm both the authority and obligation of wildlife officers to cite offenders.
The Draft Wolf Plan Should Prioritize Recovery, Conservation and Management Actions and Prioritize Securing Funds from State and Federal Sources for Implementation.
The Draft Wolf Plan has many laudable goals and strategies for conserving and managing wolves here in California. However, the Draft Wolf Plan does not detail specific priorities. It’s imperative that the Department clearly delineate priority actions for implementing the actions detailed in the Draft Wolf Plan.
Priority number one: Secure funding specifically to create and implement a Department wolf program with adequate staffing levels, appropriate training on livestock depredation investigations, and resources to run such program. This program should be equipped to provide
31 http://www.endangeredspecieshandbook.org/legislation_endangered.php 32 Uniform Bail and Penalty Schedules, Judicial Council of California, July 2011, page 121.
20 information and on-going support on the use of proactive tools and strategies available to ranchers for reducing conflicts between livestock and wolves. This should include having nonlethal tools available to lend to ranchers in need on a temporary basis.
Once funding is secured for a Department wolf program and personnel has been hired and trained for such program, the next highest priority for the Department is to ensure that at least one member of each known wolf pack should be captured and outfitted with a GPS-enabled collar so that location data can be used to inform outreach efforts, especially within the ranching community. The Department should expedite establishing Depredation Prevention Agreements with interested and willing livestock producers, which will include nondisclosure agreements to ensure that wolf location data is not inappropriately shared. (See subsequent section for additional details on this subject.)
The Department should work with the Department of Finance and other necessary entities to establish a fund to provide compensation for livestock depredations; this will go a long way to promoting goodwill among the livestock community critical to ensuring long-term wolf recovery.
The Plan Must Include a Comprehensive Plan of Action for Public Education Aimed at Recipients of Wolf-Location Information.
In addition to including a copy of the written nondisclosure agreement that wolf-location information recipients will be required to sign and adhere to, we note a specific, essential need for educational efforts by the Department to recipients of radio-collar information. Recipients must receive educational information about wolf behavior, biology and ecology, appropriate conduct around wolves and legal requirements in advance of their receipt of such sensitive information, and on an ongoing basis.
At the January 21, 2016 public meeting the Department held in Yreka, nearly 300 individuals attended, 37 of which provided oral testimony at the meeting. Much of the testimony from local residents expressed anger, resentment and fear – a desire to not have wolves in California, a disregard for state and federal law protecting wolves, and utter misinformation on what degree of threat wolves could pose to livestock or to human safety. At the same time, several spoke of the need to get radio-collars on wolves and to provide wolf location information to area ranchers.
We agree radio-collar information is important, to help the Department monitor wolves, detect if wolves have been illegally killed, and to help ranchers know when to implement or ramp up use of nonlethal conflict deterrents. Stakeholders in the planning process agreed this was important -- and conservation group stakeholders expressed strongly the need for that disclosure to remain confidential to recipients, to simultaneously be as protective of wolves as possible. Yet the wolf plan refers only vaguely to a confidentiality requirement and does nothing to address how to best ensure those in receipt of the information will not themselves become a source of harm to wolves.
21
Conservation groups and the public aren’t likely to support giving wolf location information to people who hate federal and state government, hate and fear wolves based on inaccurate information and cultural-based beliefs, and are unwilling to follow the law. And the Department has a responsibility to take all steps necessary to ensure that those in receipt of wolf-location information will adhere to the confidentiality requirement, follow the law and have as accurate as possible an understanding regarding wolf biology, behavior and ecology. For the Department to have public support for limited disclosure of wolf location information, it must develop a comprehensive plan of action, described in the Plan, for public education in areas where wolves are likely to return and specifically aimed at those individuals who will be recipients of wolf- location information.
The Draft Plan Lacks Key Information Referenced in the Draft Plan and/or which was Discussed and Intended by SWG Members to be Included in the Plan.
While the Draft Plan has a wide variety of critical provisions that will guide the state’s conservation and management of wolves into the future, it also lacks some key information that was discussed and intended by SWG members to be included in the Plan. This includes the following:
- As indicated in a separate section of our comments, Depredation Investigations Protocols should be clearly articulated and included in the Wolf Plan. - Both Wolf-Livestock and Wolf-Ungulate Conflict Management Strategies, as referenced in Table G.2c on Phase 2 Conservation Actions/Options in Appendix G are missing from the Draft Wolf Plan and should be written up and included. (Chapter 6 of the Draft Wolf Plan also states in its opening paragraph that the chapter will conclude with a discussion of the tools and strategies available for managing wolf-ungulate interactions in California, but no such discussion is included.) - The Livestock Depredation Protocol that is available on the Department’s wolf web page should also be contained within the Plan, with the understanding that it may evolve over time as we learn more about how best to address wolf-livestock conflicts in California. - List of “Priority Counties” for payment for presence and any other components relevant to these counties should be defined. We suggest Siskiyou, Modoc, Shasta and Lassen be included as Priority Counties at minimum. - More information is needed on the charge and structure of the Local/County Advisory Groups and the Plan should specify that a Statewide Advisory Group will also be established, with its charge and structure also described within the Plan. - A specific budget for the Department’s Wolf Program, including start-up and annual operating costs associated with the program. - A collaring plan and confidentiality agreement for wolf location data that will be shared by the Department with any outside interests. - An outline for how the Department will gather and use information in an adaptive management framework to undertake any future updates to the Wolf Conservation Plan and the required 5-year status review under CESA.
22
Wolves, Coyotes and Bears Should Not Be Killed To Conserve Wild Ungulate Populations
Proposed triggers for strategies for addressing any future impacts wolves – or coyotes or bears – might have on CA’s wild ungulate population allude to unspecified authority, are counter to science and do not comport with modern understandings of the ecological importance of predators. CDFW offers no peer reviewed science to buttress this approach. If anything recent research has demonstrated that killing coyotes and other predators to boost ungulate populations is questionable at best and may even be counterproductive. Therefore these unsupported and scientifically questionable triggers should be removed from the Plan.
In late 2014, the Department sent an internal draft version of the Plan for peer review by outside scientists. Reviewer Dr. Cristina Eisenberg expressly stated in her comments that while wolf recolonization and recovery in California will undeniably have impacts on ungulates, “the strengths of these impacts are impossible to fully predict” and she does “not expect that wolf predation on elk will be as much of an issue as predicted in the Plan.” Dr. Eisenberg continued, “Relocation of wolves subsequent to a reduction of allocated big game tags is not based on science, it is based on natural resources management economics. It is inappropriate to apply such an economic approach to a wolf population that is in the early stages of becoming established. It risks scapegoating wolves further, and this could have negative impacts on human perception of wolves.” Finally, she noted that, “[l]ethal control of wolves to promote elk and other prey species population growth . . . is unacceptable. Other strategies need to be implemented, such as ungulate or wolf translocation. This opens the door for lethal take without sideboards and scapegoats the wolf in a system in which predator-prey relationships will be highly complex ecologically.”33
Recent decline in elk in Montana's Bitterroot Valley was at first attributed to wolf predation. What the Montana Department of Fish, Wildlife and Parks (MFWP) discovered, however, is that the primary predator was mountain lion, not wolves. But the action that precipitated the original decline was too generous an issuance of hunting cow tags and thus human hunting was a major factor in the original decline.34
A report issued by MFWP this year surveying elk in Management Unit 313 demonstrates the effect that hunting is having on elk near Yellowstone National Park.35 The final chart in the report (Fig. 3) shows six-point bulls declining in numbers, which represents the impact of hunting outside of the Park. Decline in bulls is likely affecting the overall productivity of the elk herds. While elk herds within Yellowstone have declined over the years since the reintroduction of wolves (Fig. 2), most observers think the 19,000 elk that existed in the park prior to wolves was far too many and that the elk numbers there today are far more sustainable.
The Draft Wolf Plan acknowledges elsewhere how a decline in ungulate populations sometimes had a beneficial effect on vegetation, yet the Department and the Plan clearly consider a decline
33 Eisenberg, Dr. Cristina. California Wolf Plan Peer Review comment letter to CDFW, January 2015. 34 Perry Backus. Solving the Bitterroot Elk Mystery. How biologists and local volunteers finally figured out what was reducing the popular Ravalli County elk population. Montana Outdoors, Nov-Dec 2014. http://fwp.mt.gov/mtoutdoors/HTML/articles/2014/BitterrootElkResults.htm#.VruPem_2aM9 35 Montana Fish, Wildlife and Parks. 2016. Winter 2016 Hunting District 313 Elk Survey (Gardiner to 6- mile Creek). Prepared by MFWP biologist Karen Loveless.
23 in elk to be a "negative", as expressed in Chapter 6 and in the coyote/bear/wolf management strategies enumerated in that chapter (and in the tables in Appendix G) that would be triggered by specified declines in ungulate population numbers and/or ratios. In reality, a decline in elk herbivory pressure could be beneficial to many other species--assuming that elk numbers are high enough to be having an impact--which one can't tell from the Department’s own documentation. Studies cited elsewhere in the Plan documented higher song bird nesting populations where elk herbivory on willows declined. More willows can also result in greater beaver colonization -- which in California would be a real advantage as beaver impoundments would aid in keeping water flows during drought periods. It would also help endangered species of salmon and trout. The Plan does acknowledge that wolves might affect coyote numbers and cause other changes such as an increase in jackrabbit or higher fox survival. Yet the Plan acts as though these changes are not important if elk numbers decline.
Regarding the Plan’s reliance on specific cow/calf ratios as triggers, it is essential for the Plan to note that declines in elk cow/calf ratios usually self-correct to some degree if given time. In both the Bitterroot Valley example mentioned above, as well as in Yellowstone National Park, elk numbers declined due to predators including wolves as well as human hunting (hunting outside of Yellowstone), but after a period of time -- five to seven years -- their numbers began to recover. So the time frame is important. What may seem like a one-way decline may be more of an oscillation. The resulting elk herd is healthier with a higher proportion of reproductive age cows.
The Plan at pp. 104-105 of Chapter 6 contains some discussion of how weather/climate would affect elk and deer. The discussion is focused largely on how climate change could affect the abundance, distribution and structure of natural plant communities on which deer and elk depend for browse, and how that might affect deer and elk. However, the Plan fails to discuss the impacts of drought. This is a remarkable omission given the state of extreme drought that currently exists across much of the western United States and quite notably in California. Productivity declines significantly in drought. In Yellowstone for instance, in 1989, severe drought caused 1/3 of the elk herd to die off due to starvation (this was before wolves were present). Drought will also make elk and deer vulnerable to wolf predation, and though mortality would then in a sense be compensatory, we see a dangerous trend in this Plan which would instead blame wolves for the decline.
The Plan’s proposed action strategies for how to address an elk decline are extremely troubling and based on a pro-hunting philosophy, as opposed to being approached as a science issue. The first step should be to eliminate all human mortality, i.e. hunting pressure, and to let elk populations find their own balance. Wolves should not be killed merely to increase elk for hunters to kill. Killing wolves to increase elk for human hunters is a strategy based in philosophy, not a science-based ecosystem approach.
24
Threats To Wolves From Illegal Killing Due To Mistaken Identification As Coyotes Have Not Been Addressed.
In her peer review comment letter regarding the Draft Plan, Dr. Cristina Eisenberg emphasized the conservation threat to wolves of killings due to mistaken identity as coyotes, and urged that “coyote hunting be eliminated in California, in order to enable wolf conservation to proceed.”36
Dr. Eisenberg’s call to action is well justified and the Draft Plan’s failure to discuss this conservation threat to wolves is incomprehensible. Chapter 9 addresses Wolf Conservation and contains a discussion of threats to wolves including human-caused mortality, yet lacks any discussion whatsoever of human-caused mortality of wolves due to mistaken identification nor proposes any strategies to address this conservation threat. The environmental caucus of the SWG submitted to the Department extensive comments, proposed text and literature citations on this topic more than a year ago, yet none appear in the Draft Plan. The sole statement in the Draft Wolf Plan regarding such killings appears on p. 137, footnote 44, as a token mention of the radio-collared wolf from Wyoming which dispersed to Arizona, was named ‘Echo” by schoolchildren in a nationwide naming contest, but was then killed two months later by a hunter claiming to have mistaken the animal for a coyote.
The fact is, state and federal officials have reported wolves being shot mistakenly as coyotes in all parts of the country where wolves are returning. Environmental caucus comments sent to the Department more than a year ago included the following:
“It is essential for the safety of this state- and -federally-protected species that members of the public in California know how to distinguish a wolf from other canids. In many states where wolves are starting to return, lone dispersing wolves have been shot by hunters or landowners who stated they thought the animal they were shooting was a coyote. Between 1980-2014, 56 instances of wolves dispersing to areas outside of core recovery areas have been documented; 36 of the animals were shot and killed, another 12 found dead or killed in another manner and the fate of 8 are unknown (Weiss et al. 2014).37 Of the 36 that were shot and killed, in 11 instances the shooter expressly indicated thinking it was a coyote (Weiss et al. 2014). In late December 2014, a radio-collared wolf that had dispersed nearly 500 miles from Wyoming into Utah was killed by a hunter who said he thought it was a coyote.38 In California, as in all states, it is imperative that hunters be certain of the identity of their target before pulling the trigger; in the case of wolves, it is illegal to kill an endangered species and there are penalties, including the potential of jail-time and fines, for violating the law.”
In the 15 months since the release of the report by Weiss et al. until now, at least an additional four instances have been reported of wolves shot by hunters claiming they thought the animals were coyotes. These include the killing of the wolf known as Echo, who weighed 110 pounds; a wolf killed in Oregon in the fall of 2015, and two wolves killed in Iowa in December 2015,
36 Id. 37 Weiss et al. 2014. 38 http://www.sltrib.com/news/1999741-155/utah-hunter-kills-wolf-near-beaver
25 which weighed, respectively, 103 and 98 pounds.39 In addition, in North Carolina, at least five red wolves were killed by hunters in 2012 engaged in night-spotlight-hunting of coyotes. The instances we describe here are only the ones that agencies know about. It is highly likely far more dispersing wolves have been mistaken for coyotes and killed than have been reported or discovered.40
The Department is on notice of this threat to wolves not just because conservation group SWG members included this topic in our comment letter to the Department a year ago. In early 2013, conservation groups called the Department’s attention to a California coyote-killing contest conservation groups had just learned of, which takes place annually since its inception in 2006. That contest, sponsored by Adin Supply and the Pitt River Rod and Gun Club, based in the town of Adin in Modoc County, is held on public and private lands in the northern California counties. This region is the very pathway for dispersing wolves from Oregon to enter into and reestablish in California. Starting in 2013 and continuing over the next several years, conservation groups and the public urged state and federal officials to halt this contest and others like it in California, because contest-hunts are scientifically indefensible, unethical and inhumane to coyotes, and because they create significant risk of harm or death to legally-protected wolves that may be traversing the landscape where the contests are taking place. In the two years that the administrative petition filed by conservation groups in 2012 to list the gray wolf under CESA was pending, the Department admitted it was concerned for the safety of wolves during these contests and had sent law enforcement to the area to advise participants how to distinguish between wolves and coyotes and that it was illegal to kill a federally-protected species.41 Wolves subsequently in 2014 were listed under CESA by the California Fish and Game Commission, who also in 2015 banned giving out inducements or prizes in contest hunts of nongame mammals and furbearers, however the Adin coyote contest hunt continues to take place each year in the direct path of any wolves migrating into California from Oregon.
In light of the extensive amount of documented death-by-mistaken-identification of wolves which has occurred in states across the nation, and in light of strong concerns expressed by conservation groups and members of the general public to the Department and to the Commission since at least 2013 that coyote-killing contests in northern California are not only counter to science but a conservation threat to wolves entering the state, it is stunning that the Draft Wolf Plan includes no discussion of this threat to wolf recovery and conservation nor any discussion of strategic actions the Department will take to halt or lessen this threat. It is
39 Despite being 2-3 times larger and weighing 3-4 times more than the average coyote, wolves are being shot by hunters claiming mistaken identification. 40 Illegal killing of wolves occurs for a number of reasons. One of the chief reasons is accidental killings, either through mistaken identity or when caught in traps set for other species. 76 Fed. Reg. at 26117. It is likely that most illegal killings intentional or not, are never reported to government authorities. Id. Because the killings generally occur in remote locations and the evidence is easily concealed, there are no reliable estimates of illegal killings of gray wolves. Id. 41 In its notice of findings for the gray wolf CESA listing, the Commission confirmed that “dispersing wolves and small wolf populations are inherently at risk due to . . . being killed by hunters that mistake them for coyotes” and “[Department staff] have been fearful that . . . unknown wolves that could be in California would be mistaken for a coyote and shot or harmed.” California Fish and Game Commission, Notice of Finding and Notice of Proposed Rulemaking Gray Wolf (2014).
26 imperative this omission is remedied, complete with a strategic plan of action described by the Department.
Comments Regarding The Plan’s Assessment of Wolf Taxonomy, Population Size And Genetics Issues Regarding Hybridization.
Taxonomy. Part II of the Draft Plan, at p. 16 discusses wolf taxonomy. It is good to see that CDFW considers Canis lupus at the full species level with respect to the Wolf Plan, as we feel that this view is appropriate. The Plan is also correct that the proliferation of subspecies was a historical error and the only potential subspecies relevant to California is the Mexican gray wolf, C.l.baileyi. All of the wolves that may migrate into California, from the north, are of the same species Canis lupus.
However, we would like to comment on the inclusion in the Plan of the reference to Chambers et al., 2012. The Plan is correct to point out that scientific peer reviewers have disputed the approach and conclusions of Chambers et al. 2012. The lead author on the paper was on the USFWS staff which may have led to bias. In addition, this paper was not subject to independent peer review as is expected for legitimate scientific research. Nor is it at all clear why the conclusions of Chambers et al. 2012 would be relevant to this Plan even if its conclusions were thought to be valid by the research community.
We feel that the Chambers et al. 2012 paper should not be cited in the text or listed in Table 1.1. Disputed findings that are suspected of being biased and appeared in a non-peer reviewed journal really have no place in the CDFW wolf plan. We believe the inclusion of this material only unnecessarily confuses the reader.
Population Size. Part II of the Draft Plan, at p. 148 discusses population size. We agree that “California’s wolf population will likely be connected through migration with the larger wolf metapopulation in the Pacific Northwest, which will provide important infusions of genetic variation toward population health” and that genetic bottlenecks are unlikely. However, for this to remain true it is important that Washington, Idaho and Oregon maintain a healthy and sufficient size wolf population. California should work with those states to insure healthy populations in all.
Hybridization. Part II of the Draft Plan, at p. 150, discusses concerns regarding wolf hybridization with other canid species. We disagree with Coppinger et al. 2010 that wolves’ genomes should be considered as “fixed entities”. Nothing in nature is fixed and that is particularly true about the genomes of living animal. Variation at the level of DNA occurs continually, albeit slowly, through mutation, genetic drift and hybridization. Although hybridization with domestic dogs should not be encouraged it is nothing to excessively fear. Wolves and dogs have continued to interbreed for the past 40,000 years ever since the first wolves began to spend time near Pleistocene humans on the hunting trail. And they will in the future. In fact, one of the reasons that it is proving so difficult to establish the timeline of dog domestication is that wolves and
27 dogs have continued to interbreed over the millennia. As noted in the example of Anderson et al. 2009 dogs introduced the black coat color to the wild wolf population. We do not see hybridization as a significant threat to wolves.
The Plan Should Discuss Potential Economic Benefit to Local and Regional Economies from Wolf-Related Ecotourism and Consumer Market for Predator-Friendly Livestock Products.
The Plan does not – but should -- discuss the potential economic benefit that could accrue to entire communities or regions due to wolf-related ecotourism and those visitors’ expenditures in local economies. Conservation groups submitted published literature to the Department on this topic previously, and it should be included. The Plan also should discuss the potential financial benefit to individual ranchers who adopt predator-friendly, nonlethal coexistence measures and are able to market their product as such. Many West Coast residents would be willing to pay a premium price for beef or lamb produced without wolves being killed. A complete list of all published literature we submitted to the Department one year ago can be found in Appendix C. Articles pertaining to economic benefit from recolonizing wolves are contained in that list.
Ungulate Population and Habitat Management are Important for Wolf Conservation and Recovery Efforts.
It is important that elk populations are adequately assessed before hunting tags are increased, as was proposed in the recently rescinded Elk Environmental Assessment. The Department states that most elk populations are increasing, however, Rocky Mountain elk populations may be static and systematic surveys for elk have not been implemented in northern California.
We expect CDFW to conduct comprehensive Elk and Deer Management Plans that incorporate the needs of wolves and their effects on native elk populations. Because wolves may rely on healthy populations of ungulates it is imperative that the management plans incorporate clear goals and strategies from the California Wolf Conservation Plan, specifically relating to habitat connectivity and restoration, as well accounting for increasing wolf populations when determining ungulate population thresholds that would initiate management strategies of either species.
Throughout Chapter 6, Wolf Interactions with Ungulates, the Department stresses the difficulty of determining cause-specific mortality of ungulate populations. The cause of specific mortality for elk has not been studied in California and the overall impact from black bear, coyote and mountain lion predation on elk in California is not fully known. Deer mortality is influenced by a long list of factors that are constantly changing. The wolf plan must rely on fact rather than assumptions, when contemplating the initiation of management considerations, particularly in relation to increasing hunting of other predators such as bear and coyote.
The Department should consider the negative effects of livestock grazing on public and private lands. Cattle compete with native ungulates for habitat and forage. Based on six years of field
28 monitoring by the Project to Reform Public Land Grazing in Northern California, EPIC volunteers have found the impacts of poorly managed grazing on water quality and prime ungulate habitat has resulted in degradation, fragmentation, and overgrazing of native vegetation, such western bunchgrasses, which wild ungulate populations depend on.42
The Department Should Actively Seek Out All Opportunities to Weigh in on Land Management Actions with Federal Agencies and Participate in Land Management Planning Processes.
As outlined in Chapter 8, Coordination with Other States and Federal Agencies, we agree that working with federal land management agencies is extremely important. A strong recommendation from the Department to the U.S. Forest Service and BLM to properly manage current grazing allotments on public lands would go a long way. Current management needs to change in order to improve habitat quality, accommodate for ungulate populations and to minimize overutilization. We strongly urge the Department to participate in forthcoming national forest and BLM plan revisions in the interest of all native California wildlife, including the gray wolf.
We are encouraged to see that strategies defined in Part I of the plan include increased collaboration. We urge the Department to embrace their responsibilities in wildlife management by actively participating and collaborating with the US Forest Service, BLM, US Fish and Wildlife Service and NOAA Fisheries, as all of these agencies, including the Department, are directed to work together. Working together can include participation in upcoming national forest land management plan revisions, as is suggested on page 138, and on a project level through the National Environmental Policy Act planning processes, collaborative and partnership endeavors, such as the Western Klamath Restoration Partnership, Trinity County Collaborative and Firescape Mendocino. These large-landscape collaboratives, which include the US Forest Service as a key partner, cover a vast expanse of Northern California that includes important wolf habitat. These working partnerships are addressing issues such as wildfire, cultural and prescribed burning, wildlife habitat needs and planning treatments that will greatly affect long- term management of our public lands and habitat for both wolves and their native prey species. Please also work with the Department’s Landscape Conservation Cooperative Network.
The Plan Should Describe Priorities for Protecting, Restoring and Enhancing Habitat That Would Benefit Wolves Because the State Wildlife Action Plan Identifies the Gray Wolf as a Focal Species of Conservation Strategies
We request that CDFW develop priorities to protect, restore, and enhance habitat that would benefit gray wolves. We would like to remind CDFW of the goals, targets and strategies that are outlined in the State Wildlife Management Plan (SWAP). Table 5.2-3 in the SWAP, identifies the gray wolf as a Focal Species of Conservation Strategies Developed for Conservation Targets
42 Pace, F. 2015. Project to Reform Public Land Grazing in Northern California. http://www.wildcalifornia.org/wp-content/uploads/2016/01/Annual-Report_2015_final-final.pdf
29 in the Cascades and Modoc Plateau Province, dwelling within the North Coastal mixed evergreen and montane forests. The table below contains conservation strategy categories for the two bioregions that are associated with the gray wolf in the SWAP.
Gray wolf conservation units and targets North Coastal Mixed Evergreen and Montane Western Upland Grasslands Forests Data Collection and Analysis Data Collection and Analysis Management Planning Direct Management Land Acquisition/ easement/ lease Economic Incentives Law and Policy Land Acquisition/ Easement/ Lease Outreach and Education Land Use Planning Law and Policy
The Plan Should Identify Habitat Conservation and Connectivity Priorities that will Benefit Wolf Recovery
The Draft Wolf Plan acknowledges the significance to wolves of habitat conservation and connectivity with this statement: “First and foremost, large landscapes of suitable and non- fragmented habitat capable of supporting wolves and their primary prey are needed. This priority is not dissimilar from the habitat needs of hundreds of California wildlife species and is a basic tenet in any species conservation plan.” Part I page 13. We agree.
All of the potential concerns for wolf conservation detailed in the Habitat Alteration section of Chapter 9 point to a dire need for habitat connectivity, however the Connectivity section on page 158 of this chapter is extremely limited and is verbatim of what was in the initial Draft plan presented to the SWG a year ago. None of Environmental Caucus comments were included in this version of Draft Wolf Plan which is now out for public comment.
We are perplexed as to why the Department does not recognize its own work done on connectivity. The maps below are from the 2010 Essential Habitat Connectivity Project: A Strategy for Conserving a Connected California. For ease this document can be found online at http://www.dfg.ca.gov/habcon/connectivity/
30
We ask that the Department begin developing tangible tasks and deadlines to begin establishing these essential habitat corridors, and incorporate the maps below into the wolf Plan.
The Plan’s Trophic Cascades Discussion Should Include Published Research Demonstrating Wolves’ Positive Impacts in the Western Great Lakes States.
Chapter 1 of the Draft Wolf Plan introduces the reader to essential information about wolf biology and ecology. In general, it’s well-written but the section on trophic cascades seems to do its best to downplay potential effects generated by reestablishment of wolves. In discussing what effects wolves may or may not have on their wild ungulate prey and other parts of the ecosystem, the Draft chapter gives limited examples. It should give readers a broader, more informed perspective on this topic.
For instance, research results are discussed from a study conducted in Banff National Park showing an elk population decline after wolf recolonization. A more inclusive discussion would also contain information from the Wyoming, Montana and Idaho state agencies on elk populations, which are at or above management unit objectives nearly everywhere, and hunter harvest success rates which have for several years been at all-time highs despite the presence of
31 approximately 1600 wolves across those states. A year ago, conservation group SWG members submitted information, text and citations to the Department on this very topic. None were included in the Draft Wolf Plan but the final version of the Plan should include this information. Citations for this literature are again provided to the Department in the comprehensive list found in Appendix C.
The Draft Wolf Plan’s discussion of trophic cascades cautions against assuming that any wolf- related effects on vegetation in Yellowstone National Park could occur outside of parks. A more expansive discussion would include the research from Wisconsin examining vegetative understory in non-wolf-occupied, low-wolf-occupied and high-wolf-occupied areas. Several studies showed positive vegetative responses, especially when comparing low-wolf to high-wolf occupied areas and when sufficient time elapsed, but indicated that research design was important and design factors may have negatively impacted research results. Research results suggested that trophic cascade effects exist, are subtle, require about a decade before they are apparent, do not resemble deer-free conditions, and might become more apparent over time.43
This chapter should also include a discussion of the research from Wisconsin in which researchers concluded that as distribution of Chronic Wasting Disease in deer and wolf range overlap in the future, wolf predation may suppress disease emergence or limit prevalence.44
Chapter 1 is where most readers will obtain essential information about wolves. It is imperative that information provided regarding trophic cascades effects of wolves be more broadly representative of the current science and facts on the ground.
The Plan’s Discussion on Impacts of Wolf Mortality and Wolf-Killing on Wolf Packs Should Include the Findings of a 2014 Symposium on This Very Topic.
Part II, Chapter 9, on “Wolf Conservation” includes a short discussion on pp. 144-145 regarding responses of wolves to different levels and causes of mortality, and the effect of breeder loss on pack dynamics and size. This section cites to papers by Brainerd et al. (2008), Smith et al. (2010) and Borg et al. (2014), among others. We provide in Appendix D a more comprehensive treatment of the subject, in the form of a white paper co-authored by Dr. John Marzluff and Dr. Aaron Wirsing and two of their graduate students, from the University of Washington’s School of Environmental and Forest Sciences.
The paper is a synthesis of findings presented at a symposium held October 29th, 2014 at the University of Washington and co-hosted by the Pacific Wolf Coalition and Professors Marzluff and Wirsing. The subject was “Tackling Wolf Management’s Thorniest Issue: The Ecological and Social Complexities of Lethal Control,” and the symposium consisted of presentations given by Dr. Douglas Smith, Dr. Scott Brainerd, Dr. Adrian Treves, Dr. Jeremy Bruskotter, Dr. Rob Wielgus, Dr. Donny Martorello and Carter Niemeyer. The paper provides detailed findings described by each presenter, and sums up the findings of the presenters that lethal removal can disrupt wolf pack dynamics, inhibiting recovery objectives in recolonizing populations,
43 Callan et al. 2013; Bouchard et al. 2013; Rooney et al. PowerPoint presentation. 44 Wild et al. 2011.
32 potentially increase livestock depredations, and negatively affect human attitudes towards wolves.45
The Plan’s Discussion of Human Social Tolerance for Wolves Should Discuss and Cite to Additional Sources.
Polls and Surveys. Part II, Chapter 3, on “Human Interactions and Current Perceptions of Wolves” includes a section discussing human perceptions and attitudes towards wolves (at pp. 47-50). Page 48 notes that “[r]esearchers have conducted a number of surveys to measure human attitudes towards wolves (ranging from positive to negative) or wolf restoration, to gauge public support for such activities. Most of these efforts were conducted prior to wolf restoration and very few occurred post wolf occupancy.”
One year ago, conservation group SWG members provided the Department with surveys and polls gauging public support for wolf restoration and legal protections for wolves. Only one of the surveys we provided is discussed in this section and it relates only to people’s attitudes towards wolves. None of the polls and surveys we provided which inquired about people’s attitudes regarding wolf restoration and legal protections were discussed or cited to. Most of the polls and surveys we provided were recent, all were conducted after wolf restoration and occupancy occurred in several parts of the U.S., and almost all of them were polls and surveys that gathered data expressly from people living in California, Oregon and/or Washington or all three.
The polls and surveys we provided to the Department show overwhelming support by the public for continued legal protections for wolves, a view that wolves are a part of our natural heritage, and a desire to see wolves restored in the very state where the poll/survey respondee lived.
The discussion in Chapter 3 on polls and surveys should be more broadly representative of existing polls and surveys by including those we previously submitted to the Department. Citations for this literature are again provided to the Department in the comprehensive list found in Appendix C.
Wolf Conservation and Human-Caused Mortality. Part II, Chapter 9 discusses “Wolf Conservation.” In section B, Threats to Wolf Conservation, on pp. 143-145, the chapter discusses threats from Human-Caused Mortality. Pages 143-144 relate the historical extirpation of wolves in the conterminous United States and describe the “sport harvest” of wolves and predator control, but fail to include a discussion of the lack of agency understanding of the key threat to wolf population viability, i.e., human tolerance.
Dr. Adrian Treves, who has authored more than 100 scientific articles on ecology, conservation and society, is director of the Carnivore Coexistence Lab at the University of Wisconsin- Madison. A significant portion of his work is devoted to research on human-carnivore conflicts
45 The panel discussion was also videotaped and each panelist’s presentation can be viewed and listened to at http://www.pacificwolves.org/videos/
33
and human attitudes towards carnivores. In Chapter 3, his research team’s work analyzing data from surveys taken over an 11-year period in Wisconsin is mentioned. Dr. Treves’ research results found that when protections for wolves were lifted and state-sanctioned hunting seasons instituted, tolerance for wolves decreased, demands for more wolf-killing increased, and poaching increased. His findings are downplayed by the Department as being results obtained in the early stages of wolf recovery while the population was growing and hunting instituted only a few years ago. However, the Department entirely misses the boat on the overarching message, which is that the main threat to wolf population viability – i.e., human tolerance manifested through illegal take (poaching) – is not adequately understood by any federal or state agency yet and that the management actions agencies take in the absence of understanding can have serious repercussions. Per Dr. Treves, “ The available evidence suggests delisting and legalizing or liberalizing lethal control is more likely to increase poaching which is the major threat to wolves in the USA than decrease it.” Dr. Treves’ letter to the Oregon Fish and Wildlife Commission on this critical topic can be found in Appendix B (it is the 2nd letter in the compilation of scientists’ comment letters). We urge you to read Dr. Treves’ letter to the Commission and include a discussion of this crucial topic in section B. of Chapter 9.
The Plan’s Discussion of Human Perceptions and Interactions with Wolves Should Include Discussion and Citation to New Paper Establishing that More than 50% of Attacks on Humans by Carnivores is Due to Inappropriate Conduct by Humans.
Part II, Chapter 3, on “Human Interactions and Current Perceptions of Wolves” discusses Human Safety (at pp. 43-45) and Interaction with the Public (at pp. 45-47). Both describe instances of aggression or attacks by wolves on humans and/or contexts for those interactions. One or both of these sections should cite to and include a discussion of a recently published paper which finds that about half of all well-documented reported attacks by carnivores (black bears, brown bears, mountain lions, wolves and coyotes) in Europe and North America have involved risk- enhancing human behaviors, and that prevention and information that can encourage appropriate human behavior when sharing the landscape with large carnivores is of paramount importance to reduce both dangerous human-carnivore encounters and their consequences to carnivores. A discussion of this paper would be beneficial to any reader who lives, recreates or works in landscapes where there are bears, mountain lions, coyotes and wolves in California. It provides published research demonstrating that humans can choose to take actions which are risk- enhancing or risk-reducing and that it benefits us and California’s carnivores to be thoughtful and do the latter.46
Evidence of Historical Wolf Presence in California as Indicated Languages, Tales, Practices and Ceremonies of Native Peoples Deserves a Heading Other than “Anecdotal Observations.”
In Part II, Chapter 1, Wolf Life History and Background is discussed. At pp. 20-24, historical distribution and abundance of wolves in California, museum specimens and anecdotal observations are described. We are disappointed to see included under the heading “anecdotal
46 Penteriani et al. 2016.
34 observations” the information which comes from the languages, tales, practices and ceremonies of California’s native peoples. Given the 10,000 year history of native inhabitation of California well before the arrival of European explorers, settlers, market hunters, gold rush miners and others, it seems truly and culturally inappropriate to characterize evidence from 10,000 years of culture as mere “anecdotes.” If anything should be characterized in the Plan as anecdotes, it should be the ranchers’ and hunters’ fears, perceptions, attitudes, and beliefs, given no one in California has systematic, scientific observations of wolves to make.
During the SWG meetings, the environmental caucus requested that the Plan distinguish the evidence from California tribes in a separate section. Possibly it could be entitled “Evidence from Traditional Ecological Knowledge.” If that is not an accurate characterization of evidence from language, tales, practices and ceremonies, we feel certain the Department could come up with another suitable, distinct heading.
CONCLUSION
As detailed in the comments above, our organizations greatly appreciate the Department’s open, transparent and inclusive approach to planning for wolf conservation and management in California as the species makes its return after a nearly century-long absence. CDFW has a legal obligation to manage wildlife on behalf of all citizens of California. Promoting coexistence between wolves and livestock producers will be of critical importance to the successful management of wolves in our state, and many of our organizations stand ready to assist the Department in its effort to successfully recover gray wolf populations in California.
Numerous concerns remain about various aspects of the Draft Plan, however, including but not limited to the Department’s suggested population thresholds and duration of management phases, reducing protections at state and federal levels, lack of prioritization of actions and missing information. It is our collective hope that the Department will thoroughly review these comments and thoughtfully incorporate our suggestions to make a stronger and more comprehensive Wolf Plan that will guide recovery, conservation and management of the species well into the future.
We appreciate this opportunity to provide these comments and recommendations regarding the Department’s Draft Wolf Plan. Please do not hesitate to contact any of us if you have questions about what we have provided to you.
Sincerely,
Amaroq Weiss Pamela Flick West Coast Wolf Organizer California Representative Center for Biological Diversity Defenders of Wildlife [email protected] [email protected] 707-779-9613 916-442-5746
35
/s Jessica L. Blome Jessica L. Blome Karin Vardaman Senior Staff Attorney Director, California Wolf Recovery Animal Legal Defense Fund California Wolf Center [email protected] [email protected] 641-431-0478 949-429-9950
Nick Cady Mark Rockwell Legal Director Pacific Coast Representative Cascadia Wildlands Endangered Species Coalition [email protected] [email protected] 541-434-1463 530-432-0100
Kimberly Baker Nicole Paquette Public Land Advocate Vice President, Wildlife Protection Environmental Protection Information Center (EPIC) Humane Society of the United States [email protected] [email protected] 707-822-7711 301-258-1532
Winston Thomas, PhD Damon Nagami Pacific Region Representative Senior Attorney and Director, Living with Wolves Southern California Ecosystems Project [email protected] Natural Resources Defense Council 650-533-9979 [email protected] 310-434-2300
36
Camilla Fox Edward Moreno Founder and Executive Director Policy Advocate Project Coyote Sierra Club California [email protected] [email protected] 415-945-3232 916-557-1100 x109
Brooks Fahy Bethany Cotton Executive Director Wildlife Program Director Predator Defense WildEarth Guardians [email protected] [email protected] 541-937-4261 503-327-4923
Michael J. Connor, Ph.D. California Director Joseph Vaile Western Watersheds Project Executive Director [email protected] Klamath-Siskiyou Wildlands Center 818-345-0425 [email protected] 541-488-5789
/s Nancy Warren Maureen Hackett, M.D. Nancy Warren Founder and President Executive Director Howling for Wolves National Wolfwatcher Coalition [email protected] [email protected] 612-424-3613
/s Melissa Smith Melissa Smith President and Executive Director Friends of the Wisconsin Wolf & Wildlife [email protected]
37
LITERATURE CITED
(pdfs of listed items are in Appendix E, except Gowan et al. 2015 is in Appendix D.)
Bouchard, K., Wiedenhoeft, J.E., Wydeven, A.P., and T.P. Rooney. 2013. Boreal Environment Research 18 (suppl A): 43-49.
California Fish and Game Commission, Notice of Finding and Notice of Proposed Rulemaking Gray Wolf (2014), 9, http://www.fgc.ca.gov/CESA/gwfindingslistingwarranted.pdf
Callan, R. Nibbelink, N.P., Rooney, T.P., Wiedenhoeft, J.E. and A.P. Wydeven. 2013. Recolonizing wolves trigger a trophic cascade in Wisconsin (USA). Journal of Ecology Vol. 101: 837-845.
Cassandra Profita, The Cost of Killing Washington’s Wedge Wolves, Oregon Public Broadcasting, Nov. 14, 2012, http://www.opb.org/news/blog/ecotrope/the-cost-of-killing- washingtons-wedge-wolves/
Eisenberg, Dr. Christina. California Wolf Plan Peer Review comment letter to CDFW, January 2015.
Endangered Species Handbook. 1983. Animal Welfare Institute. http://www.endangeredspecieshandbook.org/legislation_endangered.php
Estes, J.A., J. Terborgh, J.S. Brashares, M.E. Power, J. Berger, W.J. Bond, S.R. Carpenter, T.E. Essington, R.D. Holt, J.B.C. Jackson, R.J. Marquis, L. Oksanen, T. Oksanen, R.T. Paine, E.K. Pikitch, W.J. Ripple, S.A. Sandin, M. Scheffer, T.W. Schoener, J.B. Shurin, A.R.E. Sinclair, M.E. Soule, R. Virtanen, and D.A. Wardle. 2011. Trophic downgrading of Planet Earth. Science 333: 301.
Glaser, C., Romaniello, C. and K. Moskowitz. 2015. Costs and Consequences: The real price of livestock grazing on America’s public lands. Special Report. Center for Biological Diversity. http://www.biologicaldiversity.org/programs/public_lands/grazing/pdfs/CostsAndConsequences _01-2015.pdf
Gowan, C.H., Bogezi, C., Marzluff, J.M. and A J. Wirsing. 2015. Tackling Wolf Management’s Thorniest Issue: The Ecological and Social Complexities of Lethal Control. A wolf conservation discussion panel, co-hosted by Pacific Wolf Coalition and University of Washington’s School of Environmental and Forest Sciences (October 29th, 2014). 22 pp. http://www.pacificwolves.org/wp-content/uploads/2015/10/Lethal-Wolf-Removal-Panel-White- Paper-final-08.28.2015.pdf
Imbert, C., Caniglia, R., Fabbri, E., Milanesi, P., Randi, E., Serafini, M., Torretta, E. and A. Meriggi. 2016. Why do wolves eat livestock? Factors influencing wolf diet in northern Italy. Biological Conservation 195: 156-168.
38
Montana Fish, Wildlife and Parks. 2016. Winter 2016 Hunting District 313 Elk Survey (Gardiner to 6-mile Creek). Prepared by MFWP biologists Karen Loveless.
McManus, J.S., Dickman, A.J., Gaynor, D., Smuts, B.H. and D.W. MacDonald. 2014. Dead or alive? Comparing costs and benefits of lethal and nonlethal human-wildlife conflict mitigation on livestock farms. Fauna & Flora International, Oryx, 9 pp. doi: 10.1017/50030605313001610.
Oregon Wolf Conservation and Management Plan. Oregon Department of Fish and Wildlife. December 2005 and Updated 2010. http://www.dfw.state.or.us/wolves/docs/Oregon_Wolf_Conservation_and_Management_Plan_20 10.pdf
Pace, F. 2015. Project to Reform Public Land Grazing in Northern California. http://www.wildcalifornia.org/wp-content/uploads/2016/01/Annual-Report_2015_final-final.pdf
Penteriani, V., del Mar Delgado, M., Pinchera, F., Naves, J., Fernández-Gil, A., Kojola, I., Härkönen, S., Norberg, H., Frank, J., María Fedriani, J., Sahlén, V., Støen, O., Swenson, J.E., Wabakken, P., Pellegrini, M., Herrero, S. and J.V. López-Bao. 2016. Human behaviour can trigger large carnivore attacks in developed countries. Scientific Reports, 2016; 6: 20552 DOI: 10.1038/srep20552
Perry Backus. Solving the Bitterroot Elk Mystery. How biologists and local volunteers finally figured out what was reducing the popular Ravalli County elk population. Montana Outdoors, Nov-Dec 2014. http://fwp.mt.gov/mtoutdoors/HTML/articles/2014/BitterrootElkResults.htm#.VruPem_2aM9
Ripple, W.J., J.A. Estes, R.L. Beschta, C.C. Wilmers, E.G. Ritchie, M. Hebblewhite, J. Berger, B. Elmhagen, M. Letnic, M.P. Nelson, O.J. Schmitz, D.W. Smith, A.D. Wallach, and A.J. Wirsing. 2014. Status and ecological effects of the world’s largest carnivores. Science Vol. 343: 1241484.
Rooney, T.P., Callan, R., Bouchard, K. and N.P. Nibbelink. Trophic Cascades in Great Lakes Wolves. PowerPoint presentation.
Study of Fees for Grazing Livestock on Federal Lands. A Report from the Secretary of the Interior and the Secretary of Agriculture. October 21, 1977.
Uniform Bail and Penalty Schedules, Judicial Council of California, July 2011, page 121.
Urban, M. and L. Deegan, T-shirt Weather in the Arctic, New York Times, Feb. 5, 2016. http://mobile.nytimes.com/2016/02/06/opinion/t-shirt-weather-in-the-arctic.html?_r=0
Weiss, A., Greenwald, N., and C. Bradley. 2014. Making Room for Wolf Recovery. The Case for Maintaining Endangered Species Act Protections for America’s Wolves. Center for Biological Diversity. 24 pp.
39 http://www.biologicaldiversity.org/campaigns/gray_wolves/pdfs/Making_Room_for_Recovery_ web.pdf
Wild, M.A., Hobbs, N.T., Graham, M.S. and M.W. Miller. 2011. The role of predation in disease control: A comparison of selective and nonselective removal on prion disease dynamics in deer. Journal of Wildlife Diseases. 47(1): 78-93.
Wilmers, C.C. and W.M. Getz. 2005. Gray wolves as climate change buffers in Yellowstone. PLoS Biology. 3(4):e92.
Wolf Conservation and Management Plan. Wiles, GJ., Allen, H.L. and G.E. Hayes. Washington Department of Fish and Wildlife Wildlife Program. State of Washington. December 2011. http://wdfw.wa.gov/publications/00001/wdfw00001.pdf
NEWS RELEASES
ODFW news release: Fish and Wildlife Commission delists wolves statewide in split vote (4-2). http://www.dfw.state.or.us/news/2015/November/110915.asp
Center for Biological Diversity press release: Legal Challenge Filed Over Removal of Protections From Oregon's Gray Wolves http://www.biologicaldiversity.org/news/press_releases/2015/wolf-12-30-2015.html
VIDEO
Wolf conservation discussion panel on ecological and social complexities of lethal control. October 29, 2014. Co-hosted by Pacific Wolf Coalition and University of Washington’s School of Environmental and Forest Sciences. http://www.pacificwolves.org/videos/
CASELAW
Center for Biological Diversity v. FPL Group, Inc., 166 Cal. App. 4th 1349 (2008).
Berkeley v. Superior Court, 26 Cal. 3d 515, 521 (Cal. 1980);
Illinois Central Railroad Company v. Illinois 146 U.S. 387 (1892).
National Audubon Society v. Superior Court of Alpine County, 33 Cal.3d 419, 433 (1983).
Wildearth Guardians and NMWA v US Dept of Justice, Case 4:13-cv-00392-DCB, 5/30/13, p4, item 11. US District Court, Arizona, Tucson Division.
40
STATUTES AND REGULATIONS
California
California Fish & Game Code §§ 2050-2069
California Fish & Game Code § 711.7(a)
California Fish & Game Code § 1600
Section 670.1, Title 14 of the California Code of Regulations
Oregon
ORS §§ 496.171 - 996
ORS § 498.014(a)(A)
OAR 635-110-0010(8)(b)(B)
RCW § 77.04.012
Washington
WAC 232-12-297, §4.1
WAC 232-12-297, § 4.2
Federal
16 U.S.C. §1532(19)
41
Appendix A
PROPOSED REGULATORY LANGUAGE FOR REQUIREMENTS BEFORE RESORTING TO LETHAL CONTROL
Additional Criteria, in Phase II and III for Lethal Control of Wolves to Address Chronic Livestock Depredation (if existing state and federal law allow lethal control)
Lethal take to address chronic livestock depredation. CDFW may authorize its personnel or authorized agents to use lethal force on a wolf or wolves it reasonably believes are responsible for chronic depredation upon livestock where each of the conditions in sections (1) through (6) of this rule is satisfied. CDFW shall limit lethal force to the wolf or wolves it deems necessary to address the chronic depredation situation.
Conditions for Lethal Take by CDFW. CDFW’s discretionary authority for use of lethal force pursuant to this rule may be exercised if CDFW:
1. Designates an Area of Known Wolf Activity (AKWA) and upon designation timely coordinates with potentially affected livestock producers to provide information about the California Wolf Plan, wolf behavior/management/conservation, how to document and report wolf activity to CDFW including livestock depredations, nonlethal measures/ incentives /assistance for minimizing conflicts between wolves and livestock/domestic animals in the AKWA.
2. CDFW confirms an incident of depredation by a wolf or wolves.
3. Within 14 days of CDFW’s confirmation of first wolf depredation incident, designates an Area of Depredating Wolves (ADW).
4. Concurrent with designation of ADW, prepares and publicly discloses area-specific wolf- livestock conflict-deterrence plan in coordination with potentially affected parties that identify appropriate non-lethal measures most likely to be effective for the particular circumstances.
5. Confirms a total of at least 5 separate qualifying incidents of livestock depredation on separate days within the previous 3 months by the same wolf or wolves.
6. Each of the documented depredation incidents has resulted in livestock mortality or injury.
7. Issues and makes publicly-available, prior to exercise of lethal force, written determination by CDFW Director or their designee to use lethal force to address specified situation of chronic depredation, with supported findings that (a) criteria (1)-(6) above and (8)-(13) below have been met, (b) livestock producers in ADW have worked to reduce wolf-livestock conflicts and are in compliance with wolf protection laws and conditions of any harassment or take permits, (c) the situation of depredation by wolves on livestock in ADW is likely to remain chronic despite use of additional non-lethal conflict deterrence measures and (d) wolf or wolves identified by CDFW for removal are those which CDFW finds to be associated with the qualifying depredations and CDFW finds that their removal will decrease risk of chronic depredation in ADW. 8. Qualifying Contingencies and Counting Incidents. An incident of depredation is a single event resulting in the injury or death of one or more lawfully present livestock that is reported to CDFW for investigation and, upon investigation by CDFW or its agent(s), CDFW confirms to have been caused by a wolf or group of wolves.
A qualifying incident of depredation is a confirmed incident of depredation for purposes of this rule only if:
A. If the depredation is outside an AKWA or ADW, only the first confirmed depredation by a wolf or wolves counts as a qualifying depredation. As soon as a depredation by a wolf or wolves outside of an AKWA or ADW is confirmed by CDFW, the agency must immediately designate an ADW and an AKWA and take the steps described in (1)-(4) above. If additional depredations occur outside the AKWA or ADW before the agency has acted pursuant to (1)-(4), these subsequent depredations will not count as qualifying depredations.
B. If the depredation is within an AKWA or within an ADW, the landowner or lawful occupant has, at least 7 days prior to the depredation removed, treated or disposed of all intentionally placed, known or reasonably accessible unnatural attractants such as bone or carcass piles or disposal sites; and prior to and on day of depredation incident been using non-lethal measures CDFW deems appropriate to protect the specific livestock operation there.
C. After the first depredation incident, the livestock producer has applied for or already has in place a Wolf Depredation Prevention Cooperative Agreement (WDPCA).
9. Human Presence. Human presence, when used as non-lethal measures, is presence that CDFW could reasonably expect to deter wolf-livestock conflict under the circumstances and if it occurs at proximate time prior to and in an area proximate to a confirmed depredation per CDFW and indicates timely response to wolf location information in situations of potential wolf-livestock conflict.
10. Transparency and Public Disclosure. Prior to using lethal force to address chronic wolf depredation, and with adequate notice to the public, CDFW shall document and make publicly available on at least its website (a) the determinations and supported findings referenced in section (7) above (b) but with any personal information of landowners, lawful occupants or other relevant individuals redacted from public disclosure.
11. Duration of Chronic Depredation Lethal Take Authority. Chronic depredation lethal take authority expires (a) when wolf or wolves identified for lethal removal have been removed by CDFW; (b) 45 days after issuance of the take authority unless within that time period another qualifying depredation incident occurs by same wolf or wolves identified for lethal removal and non-lethal methods have continued to have been implemented; or (c) if CDFW determines wolf or wolves identified for lethal removal have left the ADW for more than just a short-term or seasonal movement outside the area’s boundary.
Appendix B
SCIENTISTS’ LETTERS TO OREGON FISH AND WILDLIFE COMMISSION
Adrian Treves, PhD Associate Professor of Environmental Studies Director of the Carnivore Coexistence Lab The University of Wisconsin–Madison. 30A Science Hall, 550 North Park Street Madison, WI 53706 [email protected] 28 October 2015
To the Oregon Fish and Wildlife Commission:
The following comments relate to the proposal to delist gray wolves in Oregon, entitled “Updated biological status review for the Gray Wolf (Canis lupus) in Oregon and evaluation of criteria to remove the Gray Wolf from the List of Endangered Species under the Oregon Endangered Species Act (Oregon Department of Fish and Wildlife (ODFW), October 9, 2015)” hereafter “ODFW Review 2015”.
I have been studying wolf-human interactions for 16 years and ecology generally for >25 years. I’ve published >50 scientific articles on ecology, conservation and human dimensions. My lab group is the only one in the world to have measured changes in individual humans’ tolerance for wolves over time and attitudes under changing policies on lethal management and delisting. We have also studied poaching (illegal take) iin several peer-reviewed scientific publications. More information about my lab and our work on wolves can be found on our webpage: http://faculty.nelson.wisc.edu/treves/.
My comments address human tolerance for wolves, illegal take, and the public trust. I restrict my comment to two points:
(1) Oregon’s delisting criteria have not been met, and (2) The main threat to wolf population viability is not adequately understood by any state or federal agency yet, therefore the expected benefits of delisting are unlikely to manifest and the likely costs are not well addressed by current regulatory mechanisms.
By Oregon law ORS 496.17, state delisting can occur if all of five conditions are met. I address the first and fifth here.
1. The species is not now (and is not likely in the foreseeable future to be) in danger of extinction in any significant portion of its range in Oregon or in danger of becoming endangered; and
5. Existing state or federal programs or regulations are adequate to protect the species and its habitat.
Comment 1. The criteria for state delisting have not been met.
The phrase “The species is not now… in danger of extinction in any significant portion of its range in Oregon“ has two implications. The first relates to historic range and the second to not being endangered.
The historic range of the wolf in Oregon was the entire state (1) as the ODFW Report 2015 correctly noted and visible in Appendix A for map of historic range in the U.S. Habitat suitability analyses for wolves confirm that prey availability and human-caused mortality are the major factors limiting wolves from recolonizing a region, e.g., (2). If one limits the geographic extent considered to be wolf range to those areas where people want wolves to live, one opens the door to illegal and otherwise unacceptable human-caused mortality determining where wolves can live. The legal and biological flaws in this line of 2 thinking have been described and rejected for federal delisting of the gray wolf (3). In simple terms, the ODFW should not define wolf range based on interest group anger or some unquantified social acceptance, because that opens the door to a form of extortion by intolerant communities, “We’ll kill wolves that move here.” Threats posed by people are something to combat.
Instead available range should be defined by the biological capacity of wolves to find what they need to reproduce in an area and the acceptable recolonization might be determined by legal standards (see below).
With this biological logic in mind, the gray wolf is currently present in less than 6% of the state’s land area now (ODFW Review 2015), approximately equivalent to Douglas County, OR. Now imagine if the 3% of Oregon’s human population in Douglas County were the only ones to benefit from the presence of an endangered species (e.g., Washington Ground Squirrel or Lower Columbia River Coho Salmon). Wouldn’t other counties’ residents demand access without extreme efforts? Currently, too few citizens have access to the benefits generated by wolves in Oregon, which include aesthetic, ecological, and uses that deplete the asset (if that depletion leaves the asset unimpaired). Furthermore, future generations of Oregonians have a right to those benefits also. That point is emphasized by the case law upholding the public trust doctrine in Oregon. Wildlife belongs to all state citizens by Oregon law as a trust asset 1. That trust obligation limits the allocation of assets such as wildlife to private interests, e.g., livestock producers demanding lethal control of wolves (1). That trust obligation also curbs the eagerness of administrative agencies to allocate assets,
“In Morse v. Department of State Lands,2 the 1979 Oregon Supreme Court remanded the director’s decision to issue a permit authorizing a fill for an airport runway extension because he failed to determine whether the public need for the project outweighed damage to public use of trust resources…” (p. 686, section 6.2) in (4)
Therefore I recommend the Commission consider all current citizens and the rights of future generations for whom the trust is held.
I recommend that ‘a significant portion of range’ be interpreted so as to defend against litigation. I recommend ‘a significant portion of range’ be defined as one of the following geographic extents: at least one breeding pair in every county or breeding pairs in a majority of counties.
Furthermore, the current population size of wolves in Oregon “As of July 2015, there were 16 known groups or packs of wolves containing a male-female pair (Table 2), and the mid-year minimum population (non-pup) was 85 wolves.” (ODFW Review 2015). A recent illegal shooting has probably lowered that number while emphasizing the role of negligent hunters in illegal take (http://www.statesmanjournal.com/story/news/2015/10/19/man-shot-and-killed-wolf-could-face- charges/74223524/ ). At a population size <85, the addition of a few extra wolf deaths in a year can stop
1 State v. McGuire, 33 P. 666 (Or. 1883) 2 Morse, 590 P.2d at 715; After Morse, the Oregon legislature amended the Submerged and Submersible Lands Act to require the director to find that the “public need” for the project outweighs harm to public rights of navigation, fishery, and recreation. OR. REV. STAT § 196.825(3) (“The director may issue a permit for a project that results in a substantial fill in an estuary for a nonwater dependent use only if the project is for a public use and would satisfy a public need that outweighs harm to navigation, fishery and recreation and if the proposed fill meets all other criteria ... [in the Act].”). 3 or reverse population growth. As the ODFW Review 2015 noted, wolves are highly susceptible to human causes of mortality and many of these mortalities go undetected and unreported (cryptic poaching). The ODFW Review 2015 reported illegal take was the leading cause of death among wolves in a small sample of recovered mortalities. For a quantitative example from another state, we estimated an average of 44% (SD 4%) of Wisconsin wolves aged >7.5 months died each year after delisting procedures began and the state regained intermittent authority for lethal control (6). The majority of those wolf deaths went undetected and nearly half of all deaths were poached wolves. If that pattern applies after delisting in Oregon, one should expect 34–41 yearlings and adult wolves to die in the year that follows. Most will go undetected. Overcoming such high mortality rates would require higher than average population growth seen in the Oregon population (Table 2, ODFW Review 2015). Chronic, undetected, human- caused mortality challenges the success of Oregon’s wolf recovery.
Moreover hopes that delisting or state authority for lethal control will reduce poaching have been fostered by a flawed analysis (7), see (1) and (6) for why it is flawed. The actual conclusion should be just the opposite, namely delisting and legal culling authority increased poaching in Wisconsin3.
In sum, the Oregon wolf population has not met the first criterion for delisting, whether measured by geographic distribution or population size.
The next comment speaks directly to the fifth requirement that, “Existing state or federal programs or regulations are adequate to protect the species”
Comment 2. The main threat to wolf population viability is not adequately understood by any state or federal agency yet, therefore the expected benefits of delisting are unlikely to manifest and the likely costs are not well addressed by current regulatory mechanisms.
The ODFW correctly identifies the major threat to wolf population viability is human tolerance manifested through illegal take (poaching) mainly, “Since human tolerance has been and remains the primary limiting factor for wolf survival, building tolerance for this species will require acceptance of the Plan’s approach to addressing wolf conservation and human conflicts.” (p. 3, ODFW Wolf Conservation and Management Plan, December 2005 and Updated 2010)” hereafter “ODFW Plan 2010”) and same sentence on p. 34 of the ODFW Review 2015. One should expect the major threat to a listed species to be well understood and abated if delisting will succeed. Unfortunately the threat is neither well understood nor abated currently. Our evidence that illegal take has not been abated comes from the section above and data on illegal take in the past as well as the likely prospect that illegal take is likely to increase as we explain below. The evidence that human tolerance is not well understood by the ODFW comes from the ODFW Review 2015 and the ODF Plan 2010.
The ODFW Plan 2010 and ODFW Review 2015 are not up-to-date on research relating to human tolerance for wolves despite 36 instances in which those documents mentioned “tolerance” or “attitude”. There are over 100 scientific, peer-reviewed articles on human attitudes to wolves (3), and >10 recent studies from the USA address what to expect in human tolerance for wolves after intervention or after policies change (3, 8-16). The ODFW Review 2015 does not cite a single one of those studies or anything by the leaders in the field, which suggests that the ODFW has not considered the scientific evidence for the major threat to Oregon wolves.
3 Please contact the author for evidence to support this assertion in a report under review. 4
Instead, the ODFW Review 2015 cites wolf biologists who have never collected human dimensions data when making a claim about human tolerance, “There are many references which relate human tolerance to successful wolf management (Mech 1995, Bangs et al. 2004, Smith 2013).” Had the ODFW reviewed the expert scientific literature rather than biologists’ opinions, they would have learned the following:
Public acceptance for lethal control has declined significantly since the 1970s and the public prefers non-lethal methods for managing wildlife. Tolerance for carnivores and inclinations to poach them are not well predicted by wealth or economic losses but rather by peer networks and social norms that foster resistance to authority and anti-establishment actions. Those inclined to poach tend to justify their actions by over-estimating how many of their neighbors and associates do so. Tolerance for bears declined when messaging was purely negative or concerns hazards posed by wildlife. Tolerance for wolves declined after delisting and legalization of lethal management, probably because people perceived the government was sending a signal that wolves have less value or illegal take will not be enforced. The implementation of lethal control did not raise tolerance for wolves after 8 years and the inauguration of public wolf-hunting did not raise tolerance for wolves after one year. Messaging that includes a sizeable component of information on benefits is more likely to raise tolerance for carnivores than messaging that focuses on costs and risks.
The available evidence suggests delisting and legalizing or liberalizing lethal control is more likely to increase poaching which is the major threat to wolves in the USA than decrease it.
Despite the latest results described above, the scientific community still does not know enough to abate poaching, which we believe is generated by intolerance. Perpetrators of poaching are poorly studied. That creates uncertainty about who would poach a wolf, under what conditions, and where. It is widely believed that the average human’s tolerance in areas inhabited by wolves will predict behaviors that harm or help wolf conservation. If that hypothesis is false, concerns with social tolerance are misplaced and attention should focus on a few perpetrators and their social networks that promote law-breaking, rather than on the general public
I conclude that state delisting might have costs that the ODFW has not anticipated and is currently ill- equipped to understand let alone abate.
Furthermore the ODP Plan 2010 is liable to lead to an increase in poorly understood take in the wake of delisting. “A delisting decision by the Commission is not expected to significantly affect the management of wolves. This is because the Wolf Plan and associated OAR’s guide the management of wolves regardless of OESA listing status, and a delisting decision would not inherently alter the management aspects of the Wolf Plan.” (ODFW Review 2015). That is unfortunate because delisting should lead to a change in management to reduce legal AND illegal killing and increase messages about the benefits of wolves to Oregon ecosystems and citizens.
Of particular concern is whether the ODFW has correctly described the future costs and benefits of its management efforts that affect wolf survival and reproduction. Lethal management raises such concerns because there has never been a rigorous scientific experiment to test if killing wolves actually prevents future wolf predation on livestock (17-19).
Also Oregon’s state delisting would presumably activate the hunting and trapping of wolves as a “special status game mammal” under ORS 496.004 (9). (While the state wolf Plan indicates that controlled take of wolves could not occur until wolves enter into Phase III, ODFW has publically indicated that the 5 population goals established in the Plan for moving into Phase III could be met as early as 2017. The Plan also advises that it is expected that wolves will have been delisted by the time Phase III management regimes and the availability of controlled take of wolves begins. With these guidelines and the timeline ODFW has indicated, controlled take of wolves will follow delisting in short order but without scientific basis.) The expectation that “controlled take of wolves would be permitted as a management response tool to assist ODFW in its wildlife management efforts“ presumes public hunting is a useful management response. Setting aside private hunters desires to hunt or revenue generation from hunting, what conservation purpose does hunting play in a population recovering from extirpation?
Reviews of this question find little or no benefit of public hunting and trapping for conserving large carnivores (20-24). Furthermore, studies of cougars suggest public hunting can exacerbate problems with domestic animal owners (25). It may seem obvious that killing a wolf in the act of chasing, biting or otherwise attacking livestock will save that animal but the vast majority of lethal management is done far from the livestock and long after an attack has occurred. Under such indirect circumstances, lethal management is not clearly effective. Consider the unsettled dispute about lethal management of Northern Rocky Mountain wolves despite twenty years of lethal management (26, 27). Another concern is that the ODFW over-states the problem of livestock depredation in the following quote, “The challenges of wolves in areas with livestock are well documented, and wolves prey on domestic animals in all parts of the world where the two coexist”. This over-states the challenge posed by livestock predation because it ignores years of evidence that a minority of wolf packs are involved in domestic animal depredations and the geographic locations of such attacks are predictable (14, 28, 29). Moreover it ignores the many non-lethal methods that are more effective than lethal control and have not had detectable side-effects and counter-productive results such as higher livestock predation.
I recommend the ODFW pay close attention to research by independent scientists with academic freedom (not USDA-WS which has a financial conflict of interest and not hunter interest groups for the same reason) who have reviewed the evidence on whether killing wolves – either through public hunting or by USDA-WS contract – will prevent livestock predation. Otherwise, and until the scientific community finds consensus on this evaluation, any such killing authorized and condoned by ODFW is not based on best science. Indeed it is being conducted in the absence of scientific justification and may be in violation of the public trust duties of the state, as mentioned previously.
In conclusion, I find (1) Oregon’s delisting criteria have not been met, and (2) The main threat to wolf population viability is not adequately understood by any state or federal agency yet, therefore the expected benefits of delisting are unlikely to manifest and the likely costs are not well addressed by current regulatory mechanisms.
Thank you for reading my comments.
Adrian Treves, PhD
Associate Professor and Director of the Carnivore Coexistence Lab at the Nelson Institue for Environmental Studies of the University of Wisconsin–Madison. 30A Science Hall, 550 North Park Street, Madison, WI 53706, [email protected] 6
Appendix A.
Blue area is the historic range of the gray wolf in the conterminous United States. Hatched gray areas are the current range of breeding pairs of wolves as of 2013. The dark polygons show relative human population density (1).
1. A. Treves et al., Predators and the public trust. Biological Reviews, (2015). 2. D. J. Mladenoff, R. G. Haight, T. A. Sickley, A. P. Wydeven, Causes and implications of species restoration in altered ecosystems. Bioscience 47, 21-31. (1997). 3. J. T. Bruskotter, J. A. Vucetich, S. Enzler, A. Treves, M. P. Nelson, Removing protections for wolves and the future of the U.S. Endangered Species Act (1973) Conservation Letters 7, 401-407 (2013). 4. M. C. Blumm et al., The Public Trust Doctrine in Forty-Five States (March 18, 2014). Lewis & Clark Law School Legal Studies Research Paper, (2014); published online EpubMarch 18, 2014 ( 5. A. P. Wydeven et al., in Recovery of Gray Wolves in the Great Lakes Region of the United States: an Endangered Species Success Story, A. P. Wydeven, T. R. Van Deelen, E. J. Heske, Eds. (Springer, New York, 2009), pp. 87-106. 6. A. Treves, J. A. Langenberg, J. V. López-Bao, M. F. Rabenhorst, Gray wolf mortality patterns in Wisconsin from 1979–2012. J. Mammal., (in review). 7. E. R. Olson et al., Pendulum swings in wolf management led to conflict, illegal kills, and a legislated wolf hunt. Conservation Letters, (2014). 8. C. Browne-Nuñez, A. Treves, D. Macfarland, Z. Voyles, C. Turng, Evaluating the potential for legalized lethal control of wolves to reduce illegal take: A mixed-methods examination of attitudes and behavioral inclinations. Biol. Conserv. 189, 59–71 (2015). 7
9. J. T. Bruskotter, R. S. Wilson, Determining where the wild things will be: using psychological theory to find tolerance for large carnivores. Conservation Letters 7, 158–165 (2014). 10. J. Hogberg, A. Treves, B. Shaw, L. Naughton-Treves, Changes in attitudes toward wolves before and after an inaugural public hunting and trapping season: early evidence from Wisconsin’s wolf range. Environ. Conserv. doi 10.1017/S037689291500017X, (2015). 11. L. Naughton-Treves, R. Grossberg, A. Treves, Paying for tolerance: The impact of livestock depredation and compensation payments on rural citizens' attitudes toward wolves. Conserv. Biol. 17, 1500-1511. (2003). 12. K. Slagle, R. Zajac, J. Bruskotter, R. Wilson, S. Prange, Building tolerance for bears: A communications experiment. The Journal of Wildlife Management 77, 863-869. (2013). 13. K. M. Slagle, J. T. Bruskotter, A. S. Singh, R. H. Schmid, Attitudes toward predator control in the United States: 1995 and 2014. J. Mammal., (in press). 14. A. Treves, K. A. Martin, A. P. Wydeven, J. E. Wiedenhoeft, Forecasting Environmental Hazards and the Application of Risk Maps to Predator Attacks on Livestock. Bioscience 61, 451-458 (2011). 15. A. Treves, L. Naughton-Treves, V. S. Shelley, Longitudinal analysis of attitudes toward wolves. Conserv. Biol. 27, 315–323 (2013). 16. A. Treves, J. T. Bruskotter, Tolerance for predatory wildlife. Science 344, 476-477 (2014). 17. A. Treves, L. Naughton-Treves, in People and Wildlife, Conflict or Coexistence?, R. Woodroffe, S. Thirgood, A. Rabinowitz, Eds. (Cambridge University Press, Cambridge, UK, 2005), pp. 86-106. 18. A. Treves, J. A. Vucetich, M. Rabenhorst, A. Corman, "An evaluation of localized wolf control efforts to prevent subsequent livestock depredation in Michigan," Natural Resources Report No. 2013-4 (Littel River Band of Ottawa Indians, 2013)). 19. A. Treves, M. Krofel, J. McManus, Preventing carnivore predation on livestock need not be a shot in the dark. (in review). 20. I. Herfindal et al., Does recreational hunting of lynx reduce depredation losses of domestic sheep? J. Wildl. Manage. 69, 1034-1042 (2005). 21. M. E. Obbard et al., Relationships among food availability, harvest, and human–bear conflict at landscape scales in Ontario, Canada. Ursus 25, 98-110 (2014). 22. M. Krofel, R. Cerne, K. Jerina, Effectiveness of wolf (Canis lupus) culling as a measure to reduce livestock depredations. Zbornik gozdarstva in lesarstva 95, 11-22 (2011). 23. A. Treves, Hunting to conserve large carnivores. J. Appl. Ecol. 46, 1350-1356 (2009). 24. J. Vucetich, M. P. Nelson, in Political Science, Comparative Politics, Political Theory. (Oxford Handbooks Online, Oxford, UK, 2014). 25. K. Peebles, R. B. Wielgus, B. T. Maletzke, M. E. Swanson, Effects of Remedial Sport Hunting on Cougar Complaints and Livestock Depredations. PLoS ONE 8, e79713 (2013). 26. E. H. Bradley et al., Effects of Wolf Removal on Livestock Depredation Recurrence and Wolf Recovery in Montana, Idaho, and Wyoming. J. Wildl. Manage. DOI: 10.1002/jwmg.948, (2015). 27. R. B. Wielgus, K. Peebles, Effects of wolf mortality on livestock depredations. PLoS One 9, e113505 (2014). 28. E. R. Olson, A. Treves, A. P. Wydeven, S. Ventura, Landscape predictors of wolf attacks on bear- hunting dogs in Wisconsin, USA. Wildl. Res. 41, 584–597 (2014). 29. A. Treves et al., Wolf depredation on domestic animals: control and compensation in Wisconsin, 1976-2000. Wildl. Soc. Bull. 30, 231-241. (2002).
October 27, 2015
Dear Commissioners,
Soon the Commission will decide whether to remove wolves from the Oregon state list of endangered species. For reasons outlined below, we urge the Commission to refrain from removing wolves from Oregon’s endangered species list at this time.
Because Oregon state law requires delisting decisions be based on the best-available science, the Oregon Department of Fish and Wildlife has made a concerted effort to perform scientific analyses to evaluate the appropriateness of removing wolves from Oregon’s endangered species list. That analysis is reported in a document entitled, Updated biological status review for the Gray Wolf (Canis lupus) in Oregon and evaluation of criteria to remove the Gray Wolf from the List of Endangered Species under the Oregon Endangered Species Act. Hereafter we refer to that document as ODFW (2015). While the analyses described in ODFW (2015) are important, those analyses are also, by themselves, an insufficient application of best-available science. A sufficient application of best- available science also requires analyses, like those reported in ODFW (2015), to be adequately vetted by the scientific community through an independent review process. To our knowledge, that vetting has not to have taken place. In particular, we are especially concerned that the extinction risk analysis and its interpretation has not been adequately vetted. This scientific vetting is especially critical because discourse arguing for state delisting is enabled only because the U.S. Congress removed wolves from the federal list of protected species in 2011. But delisting action was based entirely and overtly on political circumstances, not best-available science. That circumstance heightens the need for Oregon to offer due diligence with respect to best-available science, where the federal government has failed.
ODFW (2015) includes analyses which strongly suggests that wolves should remain listed at this time. In particular, ODFW (2015) indicates 1) that Oregon has 106,853 km2 of currently suitable range for wolves. That is, range with sufficient prey and habitat where wolf-human conflicts are relatively minimal (as indicated by road density and land uses such as agriculture and developed areas). 2) wolves currently occupy about 12,582 km2. ODFW (2015) also implies that former range of wolves (i.e., range occupied before humans drove wolves to an endangered status) would have been greater than the current suitable range. To summarize, ODFW (2015) indicates that wolves in Oregon currently occupy less than 12% of their former range and only about 12% of current suitable range. Comparing that circumstance conditions with Oregon’s Endangered Species Act provides important context for informing Oregon’s listing judgment. In particular, the Act states that an endangered species is one that is “…in danger of extinction throughout any significant portion of its range within this state.” By that standard wolves are endangered because the species remains extirpated from nearly 90% of its currently suitable range (and extirpated from an even greater proportion of the range that wolves occupied before human persecution). Oregon state law does not require wolves to occupy all of their former range. Oregon state law does not even require wolves to occupy all of the currently suitable range. However, it is untenable to think that being extirpated from nearly 90% of current suitable range (a subset of former range) would qualify the species for delisting. This comparison between the language of Oregon’s law and wolves’ circumstance in Oregon is robustly supported by considerable scholarship and judicial opinion. Some of that peer-reviewed scholarship and judicial opinion is presented in Vucetich et al. (2006); Tadano (2007); Enzler & Bruskotter (2009); Geenwald (2009); Kamel (2010); Carroll et al. (2010), Bruskotter et al. (2013). If the Commission would be interested in a more detailed account of this scholarship for itself or its constituents, we would happily provide such an account upon request. We fully understand that wolves can be a challenging species to manage. And we appreciate that delisting may seem a solution to that challenge. However, two very important considerations suggest otherwise. First, Oregon already has many tools for managing wolf- human conflicts. Vigilant and judicious use of those tools is the key to effectively managing wolf-human conflicts. That much is clearly demonstrated by the good work of the Commission and ODFW. However, it is difficult to envision how wolf-human conflicts would be more effectively managed as a result of premature delisting. Second, the consequences of acting in haste or inconsistently with principles outlined here increase the risk that other decisions pertaining to delisting and natural resource management in general would be made out of political convenience rather than principle of law and science.
For these reasons, we urge you to refrain from removing wolves from Oregon’s list endangered species at this time.
Sincerely, John A. Vucetich, Professor of Wildlife, Michigan Technological University
Jeremy T. Bruskotter, Associate Professor, School of Environment and Natural Resources, The Ohio State University
Michael Paul Nelson, Ruth H. Spaniol Chair of Renewable Resources and Professor of Environmental Ethics and Philosophy, Oregon State University
References Bruskotter, J. T., Vucetich, J. A., Enzler, S., Treves, A., & Nelson, M. P. (2014). Removing protections for wolves and the future of the US Endangered Species Act (1973). Conservation Letters, 7(4), 401-407.
Carroll, C., Vucetich, J.A., Nelson, M.P., Rohlf, D.J. & Phillips, M.K. (2010) Geography and recovery under the US Endangered Species Act. Conserv. Biol. 24, 395-403.
Enzler, S.A. & Bruskotter, J.T. (2009) Contested definitions of endangered species: the controversy regarding how to interpret the phrase “A Significant Portion a Species’ Range”. Virginia Environ. Law J. 27, 1-65.
Geenwald, D. N. (2009) Effects on species’ conservation of reinterpreting the phrase “significant portion of its range” in the US Endangered Species Act. Conserv. Biol. 23, 1374-1377.
Kamel, A. (2010) Size, biology, and culture: persistence as an indicator of significant portions of range under the Endangered Species Act. Ecol. LQ 37, 525-561.
Tadano, N.M. (2007) Piecemeal delisting: designating distinct population segments for the purpose of delisting gray wolf populations is arbitrary and capricious. Wash. L. Rev. 82 ,795.
Vucetich, J.A., Nelson, M.P. & Phillips, M.K. (2006) The normative dimension and legal meaning of endangered and recovery in the U.S. Endangered Species Act. Conserv. Biol. 20, 1383-1390.
Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 1 of 10 25 October 2015 [email protected]
To the Oregon Fish and Wildlife Commission:
I am submitting these comments regarding the ODFW gray wolf biological status review
(ODFW 2015). I am a professional quantitative ecologist and principal scientist with the Wild Nature Institute. I have a Bachelor’s degree in Anthropology from University of California, Santa Barbara, a Master’s degree in Wildlife Natural Resource Management from Humboldt State University, and a PhD in Biological Sciences from Dartmouth College. I am an expert population biologist who has co-authored two population viability analyses (PVA) for the U.S.
Fish and Wildlife Service:
1. N. Nur, R.W. Bradley, D.E. Lee, P.M. Warzybok, and J. Jahncke. 2013. Population Viability Analysis of Western Gulls on the Farallon Islands in relation to potential mortality due to proposed house mouse eradication. Report to the National Fish and Wildlife Foundation and the US Fish and Wildlife Service. PRBO Conservation Science, Petaluma, California.
2. N. Nur, D.E. Lee, R.W. Bradley, P.M. Warzybok, and J. Jahncke. 2011.
Population Viability Analysis of Cassin’s Auklets on the Farallon Islands in relation to environmental variability and management actions. Report to the National Fish and Wildlife Foundation and the US Fish and Wildlife Service. PRBO Conservation Science, Petaluma, California.
I co-authored a comprehensive review of demography and population dynamic models (including PVA) that was part of the California Current Seabird Management Plan for U.S. Fish and Wildlife Service:
N. Nur and D. E. Lee. 2003. Demography and Population Dynamic Models as a Cornerstone of Seabird Conservation and Management in the California Current. in California Current System Seabird Conservation Plan (eds. W.J. Sydeman, K. Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 2 of 10 25 October 2015 [email protected]
Mills and P. Hodum). Report to the US Fish and Wildlife Service. PRBO Conservation Science, Stinson Beach, California.
Eight, relevant, peer-reviewed scientific articles that I have had published from my research include the following:
1. D.E. Lee, J. Bettaso, M.L. Bond, R.W. Bradley, J. Tietz, and P.M. Warzybok. 2012. Growth, age at maturity, and age-specific survival of the Arboreal Salamander (Aneides lugubris) on Southeast Farallon Island, California. Journal
of Herpetology 46:64-71.
2. D.E. Lee, R.W. Bradley, and P.M. Warzybok. 2012. Recruitment of Cassin’s Auklet (Ptychoramphus aleuticus): Individual age and parental age effects. Auk 129:1-9.
3. D.E. Lee. 2011. Effects of environmental variability and breeding experience on Northern Elephant Seal demography. Journal of Mammalogy 92:517-526.
4. A.C. Brown, D.E. Lee, R.W. Bradley, and S. Anderson. 2010. Dynamics of White Shark predation on pinnipeds in California: effects of prey abundance. Copeia 2010 No. 2:232-238.
5. D.E. Lee and W.J. Sydeman. 2009. North Pacific climate mediates offspring sex ratios in Northern Elephant Seals. Journal of Mammalogy 90:1-8.
6. D.E. Lee, C. Abraham, P.M. Warzybok, R.W. Bradley and W. J. Sydeman. 2008. Age-specific survival, breeding success, and recruitment in Common Murres
(Uria aalge) of the California Current System. Auk 125:316-325.
7. D.E. Lee, N. Nur, and W.J. Sydeman. 2007. Climate and demography of the planktivorous Cassin’s Auklet Ptychoramphus aleuticus off northern California: implications for population change. Journal of Animal Ecology 76: 337–347. Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 3 of 10 25 October 2015 [email protected]
8. S.F. Railsback, B.C. Harvey, R.R. Lamberson, D.E. Lee, N.J. Claasen, and S. Yoshihara. 2001. Population-level analysis and validation of an individual-based Cutthroat Trout model. Natural Resource Modeling 15:83-110.
I have also acted as an independent consultant offering expert advice on questions of population management and population viability for management authorities and stakeholders involved in the multi-national Action Plan under the Agreement on the Conservation of Albatrosses and Petrels.
As part of my PhD work at Dartmouth College, I conducted a PVA to explore metapopulation dynamics of giraffe in a fragmented ecosystem in Tanzania:
D.E. Lee. 2015. Demography of Giraffe in the Fragmented Tarangire Ecosystem. PhD Dissertation. Dartmouth College.
My expertise has mostly focused on seabirds and other marine predators, in addition to giraffe, but the mathematics and the biological concepts relevant to PVA are universal and well- established. The universality of the concepts is apparent in the variety of taxa population biologists like me are able to apply our expertise to. For example, my work has encompassed taxa as diverse as cutthroat trout, woodrats, mice, seabirds, seals, salamanders, spotted owls, and giraffes.
I have examined the Oregon wolf PVA and found that details of the model’s construction are vague or confused about fundamental aspects of the model, and some outputs seem to disagree with conclusions in the text. The model includes many relevant factors important to wolf population dynamics, but excludes or underestimates others such that I believe that the PVA as it was used is too simplistic and lacks sufficient detail of important demographic processes to realistically estimate probabilities of “ conservation failure” or “biological extinction” over time. Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 4 of 10 25 October 2015 [email protected]
It is my expert opinion that the existing PVA is fundamentally flawed and does not provide an adequate or realistic assessment of the Oregon wolf population to meet Criterion 1 or 2 or 4, therefore the delisting requirements are not supported by the results of the PVA as it was performed.
My primary concerns with the Oregon wolf PVA are:
1. The base model seems to produce unrealistically stable and high population growth.
2. Density-dependent survival and reproduction are not included.
3. Dispersal and territory establishment are poorly modeled.
4. Environmental and Demographic stochasticity were not explained clearly enough to convince me that the model was properly constructed.
5. Environmental stochasticity was poorly modeled.
6. Impacts of human-caused mortality were downplayed.
7. Sensitivity analyses were insufficient.
1) The base model seems to produce unrealistically stable and high population growth. Perhaps due to unrealistically high estimates of vital rates, or due to unrealistic levels of vital rate variability or covariances of vital rate variability (see below), the population growth rate of the base model is unrealistically high and stable. Page 16 of Appendix B says, “Using our baseline model, simulated wolf populations increased an average of 7% (λ = 1.07 ± 0.17 SD) per year.” This high growth rate (λ = finite rate of population growth) and its variation are comparable to recent estimates from three populations of wolves over 10 years in the northern Rocky Mountains (Gude et al. 2011). However, a recent meta-analysis of three protected and circumscribed populations monitored over 28–56 years showed population growth rates were very close to λ = 1.0, with much greater variation (SD = 0.33 to 0.51) than the Oregon wolf Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 5 of 10 25 October 2015 [email protected]
PVA described (Mech and Fieberg 2015). A summary in Fuller et al. (2003) of 19 exploited
(hunted) wolf populations monitored for 2–9 years described the average finite population growth rate as λ = 0.995 ± 0.21 SD. This leads me to believe that the Oregon wolf PVA underestimated the risk of conservation failure and biological extinction due to structural issues in the model, or due to underestimates of variability or covariation in vital rates.
2) Density dependence in survival, reproduction, and dispersal success should have been included in the model structure. What the PVA authors called density dependence was actually a simply calculated carrying capacity, or theoretical maximum wolf population size, given the current elk population, but was not in any way a realistic modeling of density dependent effects on the growing wolf population. Furthermore, wolf carrying capacity was computed in the PVA using summer elk range, when winter range, the period of greatest food limitation and the greatest limitation on elk spatial distribution, is the more realistic and conservative period during which to estimate carrying capacity.
True density-dependent effects would have recognized the documented cumulative effects of an increasing or decreasing wolf population on vital rates of survival, reproduction, and dispersal and territory establishment. It has long been known that intraspecific competition related to territoriality seems to regulate wolf density below that predicted by food availability (Stenlund 1955; Pimlott 1967, 1970; Cariappa et al. 2011). Without true density dependence in vital rates, the Oregon wolf PVA assumes wolf vital rates are the same whether wolf habitat is nearly empty of wolves, or when wolves have nearly filled all the habitat. That true density Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 6 of 10 25 October 2015 [email protected] dependence affects wolf populations was well demonstrated in Cubaynes et al. (2014) where adult survival decreased as wolf density increased, independent of prey density in the area (see
Fig. 3 from Cubaynes et al. 2014, depicted here).
3) Dispersal and territory establishment should have been modeled as a spatially explicit process using a similar spatial simulation as was used for emigration, combined with the habitat model supplied in Appendix A. The PVA uses simple probabilistic rates of dispersal and successful territory establishment. This is unrealistic given that wolves occupy exclusive, defended territories in explicit spatial arrangements, so new territories cannot be established where one already exists (Fuller et al. 2003). This relates also to the unrealistic density dependence mentioned above. Also, wolves dispersing through non-habitat will not have the same survival as wolves dispersing through suitable wolf habitat. A more realistic dispersal process would use the existing wolf habitat map and established wolf territories, keep track of additional territories as the PVA simulation progresses, and when a dispersing individual ends up in an occupied area, it must disperse again until it ends up out of the state, or in unoccupied habitat. Additionally, when wolves are travelling through non-habitat, their survival rates Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 7 of 10 25 October 2015 [email protected] should be lowered to reflect this reality. Human-caused mortality also should be increased when wolves dispersed through non-habitat. Finally, dispersal and territory establishment should have included an environmental stochasticity component.
4) Environmental and demographic stochasticity are two of the most important aspects of population viability analyses, but environmental and demographic stochasticity were poorly described, and even the authors of the Oregon wolf PVA seem confused about this topic.
Appendix B states, “We incorporated environmental stochasticity in our model by randomly drawing vital rate values from a uniform distribution with a predefined mean and standard deviation at each time step of the simulation.” What this describes is not environmental stochasticity, this is demographic stochasticity, as is stated in the next sentence of Appendix B, “…vital rates were applied at an individual level, which inherently incorporated demographic stochasticity into our model.” This confusion over demographic and environmental stochasticity is very disturbing. Nevertheless, we can establish that some level of individual demographic stochasticity is included in the model, but the authors of the PVA are unclear about the details. Drawing from a uniform distribution means all values between the lower and upper boundaries are equally likely to be selected. The authors say the values for vital rates were “from a uniform distribution with a predefined mean and standard deviation”, but this is somewhat nonsensical. What I think they mean is that they drew from a uniform distribution where the interval’s lower and upper boundaries were defined by the estimate of the vital rate’s mean, plus and minus 2 SD, however in Table 1 they say,” Values used at each time step of the analysis were randomly drawn from a uniform distribution within the specified standard deviation (SD).” So I am confused about a fundamental aspect of the PVA’s construction regarding demographic stochasticity. This is a critical point as defining the uniform distribution as the vital rate’s mean ± 1SD would make demographic stochasticity much less than if the uniform distribution’s interval was defined as the vital rate’s mean ± 2SD. Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 8 of 10 25 October 2015 [email protected]
5) The effects of environmental stochasticity are included in the model as two ‘catastrophes,’ and a prey multiplier effect. The first catastrophe resulted in complete reproductive failure for that year at the pack level to simulate diseases such as canine parovirus, and occurred with an annual probability of 0.05. The second catastrophe was modeled at the population level “to represent extremely rare, range wide events that may affect wolf populations (e.g., disease, abiotic conditions, prey population crashes),” that occurred with a probability of 0.01 and resulted in a population-wide reduction in survival of 25%. These sorts of catastrophe are indeed useful to include because rare phenomena with large demographic effects are real and often have significant effects on populations. Indeed, in the PVA as constructed, these catastrophes were important effects during early years of the simulations, before population size was large enough to be resilient to catastrophes.
Unfortunately, catastrophes are not realistic proxies for true environmental stochasticity in abiotic conditions or prey availability that are typically due to stochastic annual variation in weather patterns. True environmental stochasticity would recognize that all wolf vital rates of age-class specific survival and reproduction usually co-vary among years because they are all correlated with certain weather phenomenon (such as extremely cold, wet winters) either directly, or indirectly through the weather’s effects on prey species. Environmental stochasticity should have been modeled as a population-wide, or climate zone region-wide effect whereby all demographic parameters rise or fall together according to either a documented relationship between weather and vital rates, or a relationship between weather and prey species that indirectly affects wolf demographic vital rates.
The Oregon wolf PVA did include a prey multiplier effect (page 12) as environmental stochasticity, where, “Each year of the simulation, the prey multiplier had a 1 out of 3 chance of increasing, decreasing, or remaining the same, respectively. In years the prey multiplier increased or decreased, the maximum change was restricted to 0.10.” However, this effect Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 9 of 10 25 October 2015 [email protected] seems too small, or perhaps too limited by not affecting reproduction and dispersal, to realistically simulate true environmental variation.
Several studies have documented that the wolf populations are regulated by food, as a function of prey abundance and their vulnerability to predation (Packard and Mech 1980; Keith 1983; Peterson and Page 1988; Fuller et al. 2003). Because prey condition is highly dependent on weather conditions (Mech and Peterson 2003), wolf demography is also dependent on weather (Fuller et al. 2003). “In Denali National Park, Alaska, where humans also have little effect on the wolf population, the trend in wolf numbers from 1986 through 1994 … was driven by snow depth, which influenced caribou vulnerability (Mech et al. 1998)… As snow depth and caribou vulnerability increased, adult female wolf weights also increased, followed by increased pup production and survival and decreased dispersal (Mech et al. 1998)… In the east central Superior National Forest of Minnesota…from about 1966 to 1983, the wolf population trend followed that of the white-tailed deer herd, which was related to winter snow depth. Thus snow was seen as the driving force in the wolf-deer system (Mech 1990).” From Fuller et al. (2003).
In Isle Royale National Park, wolf population growth depended mainly on the number and age structure of the prey population, although density dependence, winter severity, and catastrophic events like disease outbreaks also play important roles (Peterson and Page 1988; Peterson et al. 1998; Vucetich and Peterson 2004).
6) Human-caused mortality impacts were significant, but conclusions downplayed the effect of human-caused mortality. The section on lethal control (page 26, Appendix B) addressed the issue of legal and illegal human-caused mortality, and concluded that reasonable levels of human-caused mortality could result in conservation failure and/or biological extinction. Probability of conservation-failure increased to 0.40 and 1.00, for mean human- caused mortality rates of 0.15 and 0.25, respectively. These results highlight the importance of anthropogenic mortality to population viability of wolves. Probability of biological-extinction was relatively low for all simulations with mean human-caused mortality rates ≤ 0.15. Comments regarding the ODFW gray wolf biological status review (ODFW 2015) by Derek E. Lee page 10 of 10 25 October 2015 [email protected]
Additionally, human-caused mortality is likely to increase as the wolf population increases, possibly leading to additional density-dependent mortality. Illegal human-caused mortality has been recorded as 30–34% of total mortality (Liberg et al. 2012; Board 2012).
Oregon Legislative Assembly changed the status of wolves to “special status game mammal” under ORS 496.004 (9). Under this classification, and when in Phase III of the Wolf Plan, controlled take of wolves would be permitted as a management response tool to assist ODFW in its wildlife management efforts. This rule would effectively allow the legal killing of all wolves in excess of the conservation objective of 4 breeding pairs. Reducing the population to such a low number would undeniably result in the impairment of wolf viability in the region. A PVA scenario should be run to quantify the probability of conservation failure and extirpation under this legally permitted management action.
7) The sensitivity analyses was simplistic and insufficient in my opinion to characterize true sensitivity of demographic parameters under different scenarios of management and environmental conditions. The PVA was supposed to focus on “determining effects of key biological processes, uncertainty in model parameters, and management actions on wolf population dynamics and viability.” I recommend a more detailed and systematic sensitivity analysis where specific parameters are individually varied ± 5, 10, and 15% to determine their impact on population growth rate. Additionally, I recommend that after the model structure and parameter values and variation has been corrected as I suggested above, several realistic management and ecological scenarios be explicitly examined to document realistic probabilities of conservation failure and biological extinction.
Sincerely,
Derek E. Lee Principal Scientist Wild Nature Institute PO Box 165, Hanover, NH 03755
Grimsö Wildlife Research Station
Comments re: ODFW’s gray wolf delisting recommendation and status review
October 29th 2015
To the Oregon Fish and Wildlife Commission:
This comment concerns the document “Updated biological status review for the Gray Wolf (Canis lupus) in Oregon and evaluation of criteria to remove the Gray Wolf from the List of Endangered Species under the Oregon Endangered Species Act (Oregon Department of Fish and Wildlife (ODFW), October 9, 2015)” in particular to the Appendix B “Assessment of Population Viability of Wolves in Oregon” hereafter termed “the PVA”.
My name is Guillaume Chapron, I am Associate Professor in quantitative ecology at the Swedish University of Agricultural Sciences and my research focuses on large carnivore conservation and management, with a particular emphasis on modeling and viability analysis. I have more than a decade of experience in this field and my research has been published in the top U.S. and international peer- reviewed scientific journals (see e.g. Chapron et al. 2014. Science 346 (6216): 1517-1519, Bauer, Chapron et al. 2015. PNAS. 10.1073/pnas.1500664112 ).
I submit this comment to help the commission in meeting the requirement outlined in OR ESA that listing decisions be based on “documented and verifiable science”.
My first comment is to congratulate ODFW for providing details on the PVA and sharing the R source code of the PVA. Such openness and transparency are not so common among agencies and deserve to be praised, as they open up for the possibility of constructive criticism. My comments are the following:
1) The PVA is not statistically correct.
A PVA typically functions by running multiple stochastic (i.e. random) trajectories of a simulated population and counting the resulting number of extinct trajectories. For example, if one would simulate 1000 trajectories and obtain 137 extinct trajectories among these 1000, the extinction probability would be 13.7%. A critical part of a viability model is therefore how stochastic processes are modeled. I have reviewed the source code of the PVA written in the R language and the way stochasticity is modeled is not correct. Taking the example of survival events, stochasticity is modeled by generating a random number from a uniform
SLU, Box 7070, SE-750 07 Uppsala, Sweden tel: +46 (0)18-67 10 00 Org.nr 202100-2817 [email protected] www.slu.se Comments re: ODFW’s gray wolf delisting recommendation and status review
distribution between 0 and 1 (as I understand it, this amounts to demographic stochasticity), and then comparing that number with another number. This latter number is randomly generated from a uniform distribution with parameters (mean- SD, mean+SD) and, as I understand it, this amounts to environmental stochasticity. This approach is fundamentally wrong for two reasons. First, the breadth of the latter distribution is restrained and values lower than mean-SD and larger than mean+SD are by default impossible (which roughly means 32% of all possible values, see the “68–95–99.7 rule”, noting that excluding the lowest values will have the most severe impact on extinction risk). Second, all values are equally likely, which is typically not the case when estimating parameters from field data as one gets a normal (or bell-shaped) parameter distribution. The PVA therefore restricts possibilities of extinction and adds noise in parameters that could be more informative. The proper way to model environmental and demographic stochasticity for survival is by using a beta-binomial mixture where beta distributed values (with shape parameters obtained through the method of moments with mean and SD) are randomly generated to serve as parameters of the binomial distribution.
The same problem is also present for litter size, where the PVA uses a uniform distribution between 2 and 8. This means that litter sizes of 1 are impossible and that litter sizes of e.g. 2, 3, 4, etc till 8 are all equally likely. This approach is simply inconsistent with wolf biology. One could use a Gamma-Poisson mixture to generate stochastic integer numbers with some environmental stochasticity.
Environmental stochasticity in the PVA is in practice implemented by sampling a vector with stride of 0.01 or 0.001. However I noticed the stride was different between environmental (0.001) and demographic (0.01) stochasticity for poaching and this is also not correct.
Finally, because the model has a quite a few parameters, I believe that running 100 trajectories is not enough to get informative and converging estimates of extinction risk and 1000 trajectories would have been a minimum. I consider the points raised in this section justify the rejection of the PVA without further consideration.
2) The PVA is not properly validated.
Calibrating and validating a complex Individual Based Model is important but can also be challenging. For the OR wolf PVA this seems to have been done by comparing simulations with a time series of 5 years. I do not believe this is statistically rigorous. Modern algorithms such as Approximate Bayesian Computation with prior-posterior inference or Pattern Oriented Modeling would be more suitable here. Note that the PVA has probably quite a few weakly identifiable parameters (pairs of different parameter values giving the same model fit). Importantly, it is not because the model was published in a peer-reviewed journal that this implies the model is validated or correct (see previous point showing it is not) and I recommend the OR wolf PVA and its R source code be peer-reviewed in
2(4) Comments re: ODFW’s gray wolf delisting recommendation and status review
an open and transparent process. Finally, I would like to point to the fact that the initial population is randomly assigned across age and social classes, which suggests the population did not start at an asymptotic stage, and early oscillations of the population structure may have affected simulations and the results of the sensitivity analysis.
3) The PVA does not use realistic parameter values or scenarios.
The PVA is parameterized with a very low poaching rate. This is not in line with what has been found in other wolf or large carnivore populations. Using a hierarchical Bayesian state-space model I have found that half the mortality of wolves in Sweden was due to poaching and that two third of poaching was not observed (Liberg, Chapron, et al. 2015. Proceedings of the Royal Society B 279 (1730): 910-915). There has been several documented cases of illegal take in OR and the total number is likely higher as illegal activities are typically under- reported. The PVA also assumes that survival rates were not influenced by social status of the animal but I question whether this is realistic as some social classes are exposed to higher mortality risks by being more active in hunting large prey.
A critical assumption of the PVA is that the past is a proper representation of the future, in particular regarding human induced mortality rates. However, the PVA in this case is actually being used to make a decision making the future different from the past (delisting). Therefore, justifying delisting based on a PVA assuming that parameters will remain constant for the next 50 years is inadequate as parameters are likely to change as soon as and if delisting happens—especially if the state moves to initiate legal hunting and/or trapping of wolves. Indeed, the PVA actually documents the effect of such changes and finds that the probability of conservation failure dramatically increases with legal mortality. A proper interpretation of the actual PVA results would actually support not delisting the wolves in OR.
Another critical assumption in the PVA is the annual immigration of 3 wolves in OR. This raises two questions. First, a population is generally considered as viable when considered as a stand-alone population and not through the regular addition of individuals. Second, the persistence of this flow of immigrants is doubtful as, for example, adjacent states are attempting to dramatically reduce their wolf populations.
4) A PVA is not the appropriate tool.
The PVA completely ignores long-term viability and the ability of OR wolves to adapt to future environmental change. However, there is a substantial amount of literature of the need for populations to have a genetically effective population size of at least Ne=500 to be considered as genetically viable and a large number of viability analyses in the conservation literature have used a package called VORTEX to include genetics aspects in viability estimates. It is unfortunate the PVA ignores such aspects and this precludes using the PVA to reach conclusions
3(4) Comments re: ODFW’s gray wolf delisting recommendation and status review
on the long-term viability of OR wolves and hence meet the requirement of OR ESA.
Worth noting is that under no possibility could a population of ~85 individuals be considered as not warranting listing under the IUCN Red List, which is a globally recognized authority in assessing species extinction risks. Similarly, the Mexican wolf population is today larger than the OR wolf one but is not at all considered as recovered by Federal authorities. There appears to be little substance for ODFW to consider a population of ~85 wolves as being recovered.
ODFW finds that the wolf is not now (and is not likely in the foreseeable future to be) in danger of extinction throughout any significant portion of its range in Oregon. However, ODFW makes this statement by implicitly removing “any significant portion of its range”, as only the outcome of a non-spatial PVA is considered sufficient. The reality is that the wolf is past being in danger of extinction throughout many significant portions of its range in OR because it occupies only 12% of its suitable habitat (so is extinct in 88% of its suitable habitat). The interpretation of this section of OR ESA by ODFW is an illegitimate interpretation that implies the suitable habitat where the species has become extinct is no longer considered as part of the species range and included in recovery targets. This interpretation also runs contrary to recent scientific literature on significant portion of range.
Finally, there has been an impressive amount of research on the ecological role wolves can play in shaping ecosystems and the report by ODFW does not consider fulfilling this role as a criteria for delisting.
Based on the points raised above, I conclude that the PVA does not provide support for delisting wolves in OR.
Yours sincerely
Guillaume Chapron, PhD, Associate Professor
Grimsö Wildlife Research Station Swedish University of Agricultural Sciences SE - 73091 Riddarhyttan, Sweden
Email: [email protected]
4(4) 29 October 2015
Oregon Fish and Wildlife Commission 4034 Fairview Industrial Drive SE Salem, OR 97302 [email protected]
RE: Oregon wolf management plan and proposal to delist gray wolf
To Whom it May Concern:
We welcome the opportunity to address Oregon Department of Fish and Wildlife 's
(ODFW) proposal to remove Endangered Species (ES) protection from the State's gray wolves
(Canis lupus irremotus). The State's wolf plan emphasizes the need to employ the best peer
reviewed and most current data available.
ODFW is seeking public input on three potential options for wolf management. The first
would remove the wolf from the State's Endangered Species designation—ODFW’s preferred
option. A second option would remove the wolf from the State’s Endangered Species designation in the eastern portion of the state but retain the Endangered Species designation in the western part of the state. The third option, our preference, maintains the wolf in protected
status on the State's Endangered Species list. The decision by Oregon will be based on the
following criteria: 1) the species is not in danger of extinction in any portion of its range; 2) the
species’ reproductive potential is not in danger of failure; 3) populations are not undergoing
imminent or active deterioration of range or primary habitat; 4) over-utilization of the species or
its habitat is not occurring or likely to occur; 5) existing state or federal programs or regulations
are adequate to protect the species and its habitat. In our opinion, the five criteria for Oregon's
delisting options to be met have not been achieved and/or are based on untested assumptions at
this time.
1
Status of the Gray Wolf in Oregon
ODFW (2015) estimates there are 77 wolves in Oregon and expects to maintain this figure with four breeding pairs per annum, as a MVP (minimum viable population) size. The wolf has been in Oregon for the past eight years and occupies, by one broad-brush estimate, approximately 15% of the State, mostly in the east. However, analysis of ODFW’s own data indicates that actual areas of current wolf use comprise only 12% of the suitable wolf habitat in
Oregon (Weiss 2015). If delisted, wolves would be protected for three additional years with the exception that animals deemed a threat to livestock could be lethally controlled.
State Wolf Management Plans & Implementation
The U.S. Fish and Wildlife Service (FWS) began removal of the gray wolf from the protection of the Endangered Species Act (ESA) in 2009. The FWS, by implementing questionable management decisions, has abrogated its conservation obligations under the ESA for gray wolves (Alderman 2009, Bruskotter et al. 2014). In lieu of federal oversight, for a historically exploited species that remains extirpated from the majority of its range, management authority has been left to the states where the species is considered federally recovered. States are required to develop wolf management plans to maintain viable populations
Oregon considers their wolf population to be recovered in spite of evidence to the contrary, consisting of very low population count and habitat saturation. By law Oregon has been mandated to develop a management plan for the species. The plan Oregon has developed is of questionable merit to maintain a viable population. Unfortunately, many state wolf management plans have failed to meet the intent, as meant by Congress, for scientifically based recovery in the short or long term (Anderson 2004, Harbine 2009). State wildlife agencies,
2
which have been given broad discretion, have frequently acted to appease vocal and politically
active anti-wolf interests, traditionally members of the agricultural, hunting and property rights
constituencies who prefer low wolf populations through large harvests and liberal regulations to
cull “problem” wolves (Anderson 2004, Bergstrom 2011, Bergstrom et al. 2009, Haber 1996,
Harbine 2009). Beyond livestock interests and some ungulate focused sportsmen groups, the
preponderance of the public and stakeholders have clearly expressed their sentiments to see
government planning maintain long term viable wolf populations able to survive predictable and
stochastic factors far beyond the foreseeable future into the next century (Alderman 2009,
Anderson 2004, Harbine 2009, Kellert et al. 1996).
Conversely, several state wolf plans take a myopic view. Creel and Rotella (2010)
document many shortcomings of these recent plans:
"Rocky Mountain wolves were removed from the ESA in May, 2009. Idaho and Montana
immediately established hunting seasons with quotas equaling 20% of the regional wolf
population [combining] ...hunting and predator control...37.1% of the Northern Rocky
Mountain (NRM) wolves were killed in the first year of delisting….unprecedented for a
species to move so rapidly from the ESA to direct harvest...strong association between
human offtake [with] additives in total mortality....in North American [wolves]...".
Creel and Rotella (2010) explain that even substantially lower harvests than allowed by the
government have been detrimental to wolves. Liberal wolf harvest quotas by Idaho’s
Department of Fish and Game have killed at least 1,100 wolves. Such a magnitude of removal
has surpassed the number considered necessary to maintain the MVP, yet Idaho continues its
unabated hunting and culling leading to the conclusion that, "open hostility toward wolves is official state policy" (Harbine 2009).
3
Idaho’s actions are blatantly not focused on maintaining wolves, but it is by no means an outlier among the states with wolf management plans. The Northern Rocky Mountain states exemplify several worst case scenarios of unscientifically and politically motivated wolf management. These attitudes, enabled by some state governments, extirpated the western wolf by the 1930's. Given tolerance for legal and illegal killings of wolves, the lack of repercussions for poaching and weakly written and unenforced recovery goals, the wolf’s recovery remains uncertain in some states (Anderson 2004, Bruskotter et al. 2014, Harbine 2009) .
Recognizing the weakness of some state plan,s the judiciary has intervened on behalf of wolves. Federal courts in Wyoming and Michigan have ordered protection reinstated for gray wolves within their jurisdictions. The courts have described management in those States as being characterized by, "lack of planning" and "reckless[ness]" (Alderman 2009).
A weakness in many of these recent state management plans is that the political process and special interest groups have promoted and fostered limited recovery actions for the species.
For many years, the FWS and states have discounted the concerns expressed by many scientists about the inadequacies and perils of the government's wolf management plans (Bergstrom 2011,
Bergstrom et al. 2009, Haber 1996, Morrell 2008). Most plans remain unchanged and several are characterized by unrealistically low MVPs (Alderman 2009, Harbine 2009, Reed et al. 2003).
It is important that Oregon in its planning and implementation avoid many of these problems exhibited by other states and regions, including removing protected status too early.
Criterion 1: “Species is Not Endangered in Any Portion of its Range.”
The gray wolf in Oregon is more than “endangered” in significant portions of its range; it is in fact absent from 88% of suitable wolf habitat in the state of Oregon (Weiss 2015). Oregon's
4 suggested MVP not only is inadequate to protect the species, but if kept this low it will not serve as an effective source of dispersers to fill the rest of the suitable wolf range in Oregon. Well established principles of conservation biology hold that populations need robust numbers of individuals for long-term viability. Recovery, as defined by ODFW, appears premature considering that the gray wolf’s return to Oregon has been for an unusually brief time and that it has not in this brief time repopulated most of the suitable habitat available.
In contrast to the current situation in Oregon, the neighboring NRM had populations establishedfor 13 years before delisting, consisting of at least several hundred animals and at least 45 breeding pairs (Fallon 2008, Reed et al. 2003). These animals had been repatriated into almost ideal conditions consisting of large tracts of wilderness and sparsely human inhabited ecosystems and optimal prey, the vast Northern Range elk (Cervus elaphus) herd, which had been so unchecked that they were instrumental agents of habitat destruction. The Upper
Midwest (Minnesota, Wisconsin and Michigan) is characterized by healthy, functional and ecologically viable wolf populations of approximately 3,700 animals, which were established in a protected status for at least several decades (Bruskotter et al. 2014, Fallon 2008, Leonard and
Wayne 2008, FWS 2014). Their range includes large tracts of protected lands that stretches into
Canada. The present status of Oregon’s wolf population is significantly less secure than in these cases.
Canid populations of fewer than 100 are insufficient to sustain a long term viable population. Alaska's Alexander Archipelago subspecies (Canis lupus ligoni) suffered a major population crash, which in one year cut its size from 221 to 89 (Edwards and Noblin 2015,
Person and Larson 2013). Their fate is contingent upon whether or not the FWS lists them as protected. At Alaska's Denali National Park, the National Park Service (NPS) was forced
5 prematurely to terminate wolf harvests, also due to a precipitous population decline (Arthur
2015). The Southwest's Mexican gray wolves (Canis lupus baileyi) had been subject to significant levels of illegal take necessitating at times recapture to protect them (Povilitis el al.
2006, U.S. Fish and Wildlife Service 2008).
Use of MVP as a strict conservation tool for species has value but should not be employed as an absolute for precise population targets; the limitations of modeling and the information they are based on require that we err on the side of caution when setting population objectives for rare species (Brook et al. 2006, Vucetich et al. 2000). Very few wildlife and plant species would be considered for delisting that have populations below 100 (Brook et al. 2006,
Bergstrom et al. 2009, Morrell 2008, Thomas 1990, Traill et al. 2007, Vucetich et al. 1997,
Wabakken et al. 2001).
Criterion 2: “Species’ Reproductive Potential is Not in Danger of Failure.”
Just as the examples cited above for long-established wolf populations, the future reproductive potential of Oregon’s wolf population and its ability to increase is unknown.
Examples from some other areas/regions have demonstrated conditions where recovering populations have not increased or declined largely due to anthropogenic factors (Liberg et al.
2011, Morrell 2008, Murray et al 2015, Povilitis el al. 2006, U.S. FWS 2008, Vucetich and
Paquet 2000). At this stage of recovery for the population in Oregon, it is too early to determine or predict its future reproductive potential with any certainty as well as the key factors that may influence it.
If removal from protection occurs for Oregon wolves, potential culling/harvest will affect the species’ reproductive potential. Culling and harvest, which will likely transpire upon
6
delisting, is a documented stress factor to packs. The concomitant physiological increase in
cortisol levels hampers fecundity in surviving animals (Bryan et al. 2014). Reproductively
viable wolf populations are characterized by a stable social hierarchy in a known territory with
an adequate prey base. Culling and harvest disrupts this social behavior and structure (Borg et
al. 2015). Adult wolves that are killed are unavailable to teach hunting techniques and maintain
control over their pack structure, leading to a reduction in reproductive output and young adult
wolves that disperse prematurely, entering unknown areas, which may increase mortality
particularly from anthropogenic sources.
It is not known with certainty if the species’ reproductive potential is secure enough to maintain a viable population and how that may change if protected status is eliminated. Wolf
populations elsewhere have exhibited significant volatility due to legal take, illegal take, disease
and habitat loss (Haber 1996, Haydon et al. 2002, Liberg et al. 2012, Sparkman et al. 2011,
Wilmers et al. 2006).
Recovery and protection planning require a goal of maintaining not only a minimum
population number, but also adequate genetic diversity (Wayne and Hedrick 2011). Larger
populations can avoid the deleterious effects of inbreeding, which in certain cases can be a
precursor to extinction (Liberg et al. 2005, Peterson and Krumenaker 1989, Wayne et al. 1991).
On Isle Royale a population of over 50 wolves is now down to three, a partial result of
inbreedin g effects. Low populations of other large predators have demonstrated the impacts of
the loss of genetic diversity to population vitality (Florida panthers, Puma concolor and
cheetahs, Acinonyx jubatus) (Hedrick and Fredrickson 2010, Johnson 2010, O’Brien et al. 1985).
Wayne and Hedrick (2011) state that, "isolated populations of less than 100
individuals...have a high chance of extinction...genetic loss could be consequential...these
7
populations could lose 2.5% [genetic heterozygosity] per generation.” A generally accepted
number of 500 animals is considered an acceptable baseline estimate for a Minimum Viable
Population (MVP) to avoid inbreeding depression in the short term (Brook et al. 2006).
Preferably, the MVP would be 1,200-2,500 (Fallon 2008). Wolves in the NRM specifically
should have 1,403 individuals with a goal of 6,332 at 40 generations to better ensure a
genetically diverse, robust population to last for a century (Reed et al. 2003).
Oregon's option for removal of the gray wolf from protection at this time does not
provide safeguards for the genetic diversity of the population, which could have deleterious drift
effects and inbreeding depression, nor a large enough population for maintaining reproductive
potential in the long-term.
Criterion 3: “Populations are Not Undergoing Imminent or Active Deterioration of Range
or Primary Habitat.”
A diverse array of suitable prey and habitat exists for wolves in Oregon (Larsen and
Ripple undated), yet only about 12% of it is currently occupied by wolves (Weiss 2015).
Western Oregon is heavily urbanized with a continuously growing human population, a frequent
deterrent for many of the West's exploited wolves as well as a potential for increased cause of
mortality due to transportation infrastructure and potential for higher levels of human
disturbance. Conversely, rural Eastern Oregon is characterized by agricultural and timber
resources where those interests frequently view wolves in a negative manner, which in some
circumstances results in unregulated take (Anderson 2004, Liberg et al. 2011).
The importance of travel corridors between subpopulations of wolves in the State is also
not well known or documented at this time due to their limited number. Travel corridors are
8
considered essential for maintaining disjunct wolf populations (Haight et al. 1998). Paquet et al.
(2009) stress the significance of avoiding anthropogenic impacts to wildlife movement. Wolves
are what Paquet et al. (2009) refers to as "passage species," which "need corridors to allow
individuals to pass directly between two areas in discrete events of brief duration (e.g. dispersal
of a juvenile, seasonal migration, or moving between parts of a large home range)." Oregon is
considered to have less contiguous habitat and more patches spread out than in the NRM, which
would require wolves to cross areas of unsuitable habitats (Larsen and Ripple undated).
Essential corridors for wolf populations and habitat patches within Oregon are poorly known and
understood, because the species is still in an early stage of recovery.
Criterion 4: “Over-utilization of the Species or its Habitat is not Occurring or Likely to
Occur.”
Wolves were efficiently extirpated in the contiguous U.S. by the 1930s. FWS describes
the event as, "wolves were hunted and killed with more passion and zeal than any other animal in
U.S. history" (FWS 1998). Given our modern technology and increased network of roads even
in protected lands, anyone can access wolf habitat and cause the take of wolves. Many
individuals have access to year-round use of off-road vehicles, aircraft, drones, traps and high- powered guns. As just one example of the ability for rapid reductions, in Canada, the government permitted the taking of several hundred wolves in several days by hunters on snowmobiles with firearms, indicating the ease of making large reductions in a short period of time (Cluff 2003). Key habitat areas and components, as well as essential travel corridors and their level of protection, are poorly understood and should be addressed before delisting occurs.
9
Poaching has been estimated to comprise 30% of known mortality in some wolf populations
(Liberg et al. 2012), and roadkill as much as 11% (Fuller 1989).
Criterion 5: “Existing State or Federal Programs or Regulations are Adequate to Protect the Species and its Habitat.”
The gray wolf has only recently returned to Oregon. It still occurs in isolated populations and occupies only a fraction of its former range. Eliminating existing state protection under its current status is not justified at this stage. Setting the regulatory environment to increase “take” or modifying the current status of protected habitat when the population is at an early stage of recovery is not justified for meeting a goal of maintaining a viable population. There are many examples where wolf populations have declined or where recovery has been delayed/slowed due to a lack of or limited government protection (Bergstrom 2011, Bruskotter et al. 2014, Edwards and Noblin 2015, Liberg et al. 2011, Morrell 2008, Murray et al. 2015, Polivitis et al. 2006,
Vucetich and Paquet 2000).
Comments on the population viability analysis (PVA)
Although we have not examined the PVA in great detail, we do have a few concerns and questions about the reliability of some assumptions made about wolf vital rates, and the lack of application of the best and most recent science on these specific issues. First, it appears that the modelers assumed that current population growth rates of the Oregon wolf population, which are indicative of an expanding population filling vacant habitat, will continue indefinitely. If so, that is unrealistically optimistic. The reality for the gray wolf population of Yellowstone National
Park (YNP) after reintroduction was that density peaked at > 170 animals in the early-mid 2000s,
10
and then fell to around 100 or fewer for the last 5 years
(http://www.nps.gov/yell/learn/nature/wolfmgnt.htm). In other words, density dependent self-
regulation took control some years after establishment (demonstrated in a statistical analysis of
this population; Cubaynes et al. 2014), causing first a negative population growth rate and
ultimately an equilibrial density that was considerably lower than peak density. The latter would
constitute a more realistic model of what happens to wolf population dynamics when the habitat
fills up.
Second, we think that 88% adult survival is unrealistically high, given that the most
recent analysis of the unhunted YNP population revealed a natural mortality rate of 20% (95%
C.I. ranging from 5-50%, and with higher mortalities at higher densities when inter-pack
aggression increased). The same study also indicated an 8.4% mortality due to roadkill. To the
extent that the PVA incorporates human-caused mortality, it should not be assumed that such
mortality is largely compensatory (i.e., trades off with natural mortality, or is compensated for by
increased recruitment as a response). Any rigorous sensitivity analysis within PVA would be
remiss if it did not model the effects on wolf population growth rates of human-caused mortality
acting in an additive and even super-additive manner. Creel and Rotella (2010) found that in the
Northern Rocky Mountain (NRM) metapopulation, human offtake was indeed super-additive.
Similarly, Ausband et al. (2015) found for NRM wolf populations that not only did recruitment
not increase to compensate for human-caused mortality, but it actually decreased (again, the
super-additive effect).
Finally, poaching is likely to be a significant source of mortality, which is often overlooked or underestimated, and therefore must be modeled. Although difficult to determine, 2 recent estimates of the proportion of total mortality in wolf populations that is due to poaching
11
are 30% for Sweden (Liberg et al. 2012) and 34% for Wisconsin (Natural Resources Board
2012).
Our Recommendations
Even though we feel that the USFWS-imposed threshold for listing (relisting) of the gray wolf in each of the three neighboring NRM states—150 wolves and 15 breeding pairs sustained for at least 3 consecutive years—is unjustifiably low given the science we have briefly reviewed above, Oregon’s current population is only half that. Therefore, it would be preferable if Oregon not entertain the prospect of delisting the gray wolf at least until that landmark was achieved for
3 or more consecutive years. That should also allow the gray wolf to increase its utilization of suitable habitat in the state beyond its current 12%. In the meantime, the species should be closely monitored and information collected to aid in the continuing refinement of the state’s wolf management plan, specifically identifying and protecting key habitat and travel corridors,
developing innovative policy and guidance for agricultural interests to reduce the need for
removal/culling (Niemeyer 2012, Shivik 2006, Wielgus and Peebles 2014), and providing a
focus on maintaining a population at a level for a functional ecosystem role (Licht et al. 2010).
Cordially,
Alex Krevitz, M.A Kunak Wildlife Studies 33992 Rivercrest Rd N Coarsegold, CA
Anthony J. Giordano, Ph.D. Carnivore Biologist & Executive Director, S.P.E.C.I.E.S. Ventura, CA
12
Bradley J. Bergstrom, Ph.D. Professor of Biology Valdosta State University Valdosta, GA
Rodney L. Honeycutt, Ph.D., University Professor Chairperson of the Natural Science Division Pepperdine University Malibu, CA
Nathan S. Upham, Ph.D. NSF Postdoctoral Associate Department of Ecology and Evolutionary Biology Yale University New Haven, CT
Winston P. Smith, Ph.D. Principal Research Scientist Institute of Arctic Biology University of Alaska - Fairbanks Juneau, AK
Steven R. Sheffield, Ph.D. Associate Professor of Biology Bowie State University Bowie, MD, and Adjunct Professor College of Natural Resources and Environment Virginia Tech, National Capital Region Northern Virginia Center Falls Church, VA
References
Alderman, J.H. 2009. Crying wolf: the unlawful delisting of Northern Rocky Mountain gray
wolves from Endangered Species Act protections. Boston College Law Review
50(4):1195-1241.
Anderson, M.B. 2004. Federal delisting of the gray wolf: an Oregon perspective on the future of
13
gray wolf recovery under State endangered species acts. Vermont Journal of
Environmental Law, Vol 6:133-165.
Arthur, S. 2015. Wolf monitoring in Denali National Park and Preserve, 2014-2015. National
Park Service, Spring 2015 Report.
Ausband, D.E., C.R. Stansbury, J.L. Stenglein, J.L. Struthers, and L.P. Waits. 2015. Recruitment
in a social carnivore before and after harvest. Animal Conservation
doi:10.1111/acv.12187
Bergstrom, B.J., S. Vignieri, S.R. Sheffield, W. Sechrest, and A.A. Carlson. 2009. The northern
Rocky Mountain gray wolf is not yet recovered. BioScience 59: 991-999.
Bergstrom, B.J. 2011. Endangered wolves fall prey to politics. Science 333:1092.
Borg, B.L., S.M. Brainerd, T.J. Meier, and L.R. Prugh. 2015. Impacts of breeder loss on social
structure, reproduction and population growth in a social canid. J. of Animal Ecology
84:177-187.
Brook, B.W., L.W. Traill and C.J.A. Bradshaw. 2006. Minimum viable population sizes and
global extinction risk are unrelated. Ecology Letters 9:375-382.
Bruskotter, J.T., J.A. Vucetich, S. Enzler, A. Treves, and M.P. Nelson. 2014. Removing
protections for wolves and the future of the U.S. Endangered Species Act (1973).
Conservation Letters 7(4):401-407.
Bryan, H.M., J.E.G. Smits, L. Doren, P.C. Paquet, K.E. Wynne-Edwards, and M. Musiani.
2014. Heavily hunted wolves have higher stress and reproductive steroids than wolves
with lower hunting pressure. Functional Ecology: doi: 10.1111/1365-24355. 1-10.
Cluff, D. 2003. Personal Communication
Creel, S. and J. Rotella. 2010. Meta-analysis of relationships between human off take, total
14
mortality and population dynamics of gray wolves (Canis lupus). PLoS ONE 5 (9):1-9.
Cubaynes, S., D.R. MacNulty, D.R. Stahler, K.A. Quimby, D.W. Smith, and T. Coulson. 2014.
Density-dependent intraspecific aggression regulates survival in northern Yellowstone
wolves (Canis lupus). Journal of Animal Ecology 83:1344–1356.
Edwards, L. and R. Noblin. 2015. Imperiled wolf population on Alaska’s Prince of Wales
Island crashes. Center for Biological Diversity Press Release, dtd June 5, 2015.
Fallon, S. 2008. A petition to prepare a recovery plan under the Endangered Species Act for the
gray wolf. Natural Resources Defense Council and Defenders of Wildlife. Petition dtd
February 20, 2008. 38pp.
Fuller, T.K. 1989. Population dynamics of wolves in north-central Minnesota. Wildlife
Monographs 105:3-41.
Haber, G.C. 1996. Biological, conservation and ethical implications of exploiting and
controlling wolves. Conservation Biology 10 (4):1068-81.
Haight, R., G. D. Mladenoff and A. Wydeven. 1998. Modeling disjunct gray wolf populations in
semi-wild landscapes. Conservation Biology 12 (4):879-888.
Harbine, J.K. 2009. Gray wolves in the Northern Rockies again staring down the barrel at
hostile state management. Ecology Law Currents 36:195-204.
Haydon, D.T., M.K. Laurenson, and C. Sillero-Zubiri. 2002. Integrating epidemiology into
population viability analysis: managing the risk posed by rabies and canine distemper to
the Ethiopian wolf. Conservation Biology 16 (5):1372-1385.
Hedrick, P.W., and R. Fredrickson. 2010. Genetic rescue guidelines with examples from
Mexican wolves and Florida panthers. Conservation Genetics 11:615-626.
Johnson, W.E. 2010. Genetic restoration of the Florida panther. Science 329:1641-1645.
15
Kellert, S.R., M. Black, C.R. Rusk, and A.J. Bath. 1996. Human culture and large carnivore
conservation. Conservation Biology 10(4):977-990.
Larsen, T.E., and W.J. Ripple. Not Dated. Modeling gray wolf habitat in the Pacific Northwest,
U.S.A. Oregon State University.
Leonard, J.A. and R.K. Wayne. 2008. Native Great Lakes wolves were not recovered. Biology
Letters 4:95-98.
Liberg, O, H. Andren, H.C. Pedersen, H. Sand, D. Sejberg, P. Wabakken, M. Akesson, and S.
Bensch. 2005. A severe inbreeding depression in a wild wolf (Canis lupus) population.
Biology Letters 1:17-20.
Liberg, O., G. Chapron, P. Wabbaken, H.C. Pedersen, N.T. Hobbs, and H. Sand. 2012. Shoot,
shovel and shut up: cryptic poaching slows restoration of a large carnivore in Europe.
Proc Royal Society B. 279:910-915.
Licht, D.S., J. J. Millspaugh, K.E. Kunkel, and C.O. Kochanny. 2010. Using small populations of
wolves for ecosystem restoration and stewardship. Bioscience Vol 60(2):147-153.
Morrell, V. 2008. Wolves at the door of a more dangerous world. Science 319:890-892.
Murray, D.L., G. Bastille-Rousseau, J.R. Adams, and L.P. Waits. 2015. The challenges of red
wolf conservation and the fate of an endangered species recovery program. Conservation
Letters DOI 10.111/con12157.
Natural Resources Board. 2012. Adoption of board order wm-09012(e) relating to wolf hunting
and trapping regulations, establishment of a depredation program, and approval of a
harvest quota and permit level. Pp. … (N. R. Board, ed.). Wisconsin Department of
Natural Resources, Madison, WI.
Niemeyer, C. 2012. Suggestions for changing Wildlife Services range from new protocols to
16
outright bans. Tom Knudson. Sacramento
Bee.wwwsacbee.com/2012/05/06/4469607/suggestions-in-changing-wildlife.html
O’Brien, S.J., M.E. Roelke, L. Marker, A. Newman, C.A. Winkler, D. Meltzer, L. Colly, J.F.
Evermann, M. Bush, and D.E. Wildt. 1985. Genetic basis for species vulnerability in the
cheetah. Science 227:1428-1434.
Oregon Department of Fish and Wildlife. 2015. Wolf Management Plan. Salem, Oregon
Oregon DFW- Assessment of Population Viability of Wolves in Oregon.
Paquet, P.C., J.R. Strittholt, and N.L. Staus. 1999. Wolf reintroduction feasibility in the
Adirondack Park. Conservation Biology Institute, Corvallis, OR. 97330.
Person, D. and K. Larson. 2013. Developing a method to estimate abundance of wolves in
southeast Alaska. ADF&G Division of Wildlife Conservation.
Peterson, R.O., and R.J. Krumenaker. 1989. Wolves approach extinction on Isle Royale: a
biological and policy conundrum. George Wright Forum 6:10-15.
Polivitis, A., D.R. Parsons, M.J. Robinson, and C.D. Becker. 2006. The bureaucratically
imperiled Mexican wolf. Conservation Biology 30(4):942-945.
Reed, D.H., J.J. O'Grady, B.W. Brook, J.D. Ballou and R. Frankham. 2003. Estimates of a
minimum viable population size for vertebrates and factors influencing those estimates.
Biological Conservation 113:23-34
Ripple, W.J., R.L. Beschta, J.K. Fortin, and C.T. Robbins. 2014. Trophic cascades from
wolves to grizzly bears in Yellowstone. Journal of Animal Ecology 83:223-233
Shivik, J. 2006. Tools for the edge: what’s new for conserving carnivores. BioScience
56(3):253-259.
Sparkman, A.M., L.P. Waits, and D.L. Murray. 2011. Social and demographic effect of
17
anthropogenic mortality: a test of compensatory mortality hypothesis in the red wolf.
PloS ONE DOI: 10 1371/journal.pole.0020868.
Thomas, C.D. 1990. What do real population dynamics tell us about minimum viable
population sizes? Conservation Biology 4:324-327.
Traill, L.W., J.A. Bradshaw, and B.W. Brook. 2007. Minimum viable population size: a meta-
analysis of 30 years of published estimates. Biological Conservation 139:159-166.
U.S. Fish and Wildlife Service. 1998. Gray Wolf.
http://training.fws.gov/library/pubs/graywolf.pdf
U.S. Fish and Wildlife Service. 2008. Mexican wolf population survey complete. Press
Release.
U.S. Fish and Wildlife Service. 2014. Wolf - Western Great Lakes, current population in the
United States. USFWS – website.
Vucetich, J., and P. Paquet. 2000. The demographic population viability of Algonquin wolves.
Prepared for The Algonquin Wolf Advisory Committee. 22pp.
Vucetich, J.A., R.O. Peterson, and T.A. Waite. 1997. Effects of social structure and prey
dynamics on extinction risk in gray wolves. Conservation Biology 11(4):957-965.
Vucetich, J.A., T.A. Waite, L. Qvanemark, and S. Ibarguen. 2000. Population variability and
extinction risk. Conservation Biology 14(16):1704-1714.
Wabbaken, P., H. Sand, O. Liberg and A. Bjarvall. 2001. The recovery, distribution and
population dynamics of wolves on the Scandinavian peninsula, 1978-1998. Can J Zool:
79: 710-725.
Wayne, R.K. and P. Hedrick. 2011. Genetics and wolf conservation in the American West:
lessons and challenges. Heredity 107(1):16-19.
18
Wayne, R.K., N. Lehman, D. Girman, P.J.P. Gogan, D.A. Gilbet, K. Hansen, R.O. Peterson, U.S.
Seal, A. Eisenhawer, L.D. Mech, and R.J. Krumenaker. 1991. Conservation genetics of
the endangered Isle Royale gray wolf. Conservation Biology 5(1):41-51.
Weiss, A.E. 2105. Letter of 24 April 2015 to Oregon Fish and Wildlife Commission from
Center for Biological Diversity concerning proposal to remove gray wolf from list of
protected species in the State of Oregon.
Wielgus, R.B., and K.A. Peebles. 2014. Effects of wolf mortality on livestock depredations.
PloS ONE DOI:10.1371/journal.pone.0113505.
Wilmers, C.C., E. Post, R.O. Peterson, and J.A. Vucetick. 2006. Predator disease outbreak
modulate top-down, bottom-up and climatic effects on herbivore population dynamics.
Ecology Letters 9(4):383-389.
19
October 25, 2015
Oregon Department of Fish and Wildlife Commission 4034 Fairview Industrial Drive SE Salem, OR 97302 [email protected]
Chair Finley and Commissioners:
My name is Robert Beschta, I am emeritus professor in the Department of Forest Ecosystems and Society at Oregon State University (professional affiliation provided for informational purposes only). For more than four decades I have participated in research, teaching, and extension activities assessing the effects of land use practices on watersheds and plant communities. Much of that effort was in Oregon but more recently I have done research in Yellowstone National Park and other areas of the American West.
When wolves were extirpated from Yellowstone National Park, increased herbivory by elk soon began to impact plant communities. Over time, and over a wide range of elk densities, the park’s aspen, willow, cottonwood, alder, and a wide range of berry-producing shrubs were less able to establish and grow above the browse level of elk; tall forbs and native grasses were also impacted. As a consequence, streams eroded and incised, riparian habitat for birds and other wildlife became limited, and beaver disappeared.
After seven decades of absence, wolves were returned to the park in the mid-1990s thus completing the wild predator guild. With the return of this apex predator, changes to previously browsing-suppressed plant communities began to occur. Initially these effects were small and local but over time the effects have become more widespread. Increasingly aspen and riparian plant communities have become more robust, increasingly plants are growing above the browse level of elk, stream banks are stabilizing, more birds have habitat, and beaver are returning. These effects did not happen overnight, but have become more pronounced over the last several years. It is important to note that Yellowstone is not a unique, stand-alone experiment. Improving plant communities have also been observed in other areas of western North America where formerly extirpated wolves have returned.
Like Yellowstone, wolves were extirpated from Oregon and were absent over many decades. Elk numbers, which had been reduced to only a few thousand in the early 1900s have since increased greatly and in 2011 Oregon’s total elk numbers were 3rd highest of 11 western states (based on estimates of the Rocky Mountain Elk Foundation). And, like Yellowstone, wolves have returned. Oregon’s wolf conservation and management plan indicates “Wolves need to be managed in concert with other species and resource plans.” Most people would likely assume “other species” simply means elk. I would strongly suggest that we need to look deeper.
Deciduous woody plant communities on public lands in eastern Oregon, plant communities such as those associated with aspen and riparian areas, have experienced major declines over much of the 20th century with adverse consequences to terrestrial wildlife species as well as aquatic species, such as salmon. While outmoded livestock practices have been a major reason for this decline, herbivory by wild ungulates, principally elk, is now a significant factor in many areas and may limit recovery of degraded plant communities even if livestock impacts are minimized.
Whether the positive ecosystem effects found in Yellowstone and other areas following the return of wolves will occur in Oregon is not yet known. However, if wolves are going to be a factor in the recovery of degraded aspen stands and riparian plant communities on public lands in eastern Oregon, I would strongly indicate that delisting this keystone species is a move in the wrong direction.
Sincerely,
Robert L. Beschta
Robert L. Beschta, PhD
4005 NW Princess St.
Corvallis, OR 97330
WWW.KLAMATHCONSERVATION.ORG Klamath Center for Conservation Research PO Box 104, Orleans, CA 95556 USA
Via email to: Russ Morgan Wolf Program Coordinator Oregon Department of Fish and Wildlife 107 20th Street La Grande, OR 97850
October 28, 2015
Scientific peer review comments on Oregon Department of Fish and Wildlife Review of the Biological Status of the Gray Wolf
Thank you for your invitation to submit comments on the updated biological status review document of October 9, 2015. My research as a wildlife ecologist with the Klamath Center for Conservation Research in Orleans, California, has focused on habitat, viability, and connectivity modeling for a diverse group of threatened and endangered species ranging from large carnivores to rare and endemic plant species. I have also served on the Science and Planning Subgroup of the Mexican Wolf Recovery Team. I welcome the opportunity to use this expertise to evaluate the document.
Firstly, I wanted to commend the Oregon Department of Fish and Wildlife (ODFW) for its work over the past decade to advance wolf recovery in Oregon, and specifically on the work that went in to preparation of the biological status review document. On the whole, the document is well-written, factual, and informative. However, there are several areas where the document could be improved to better reflect current science. Although the document states that a change in status (delisting) of Oregon wolf populations will have little practical short-term effect on management of the species in the state, it is nonetheless important that any status determination reflect best available science.
The population viability analysis (PVA) completed by ODFW to support the status report provides relevant information concerning some factors effecting population status. The PVA results support the intuitive conclusion that the relatively high reproductive rate shown in many colonizing wolf populations
1
WWW.KLAMATHCONSERVATION.ORG Klamath Center for Conservation Research PO Box 104, Orleans, CA 95556 USA
make them fairly resilient to extirpation in the short term in the absence of high human-associated mortality rate (such as from hunting or lethal control programs). This conclusion can be drawn from simple deterministic PVA models. The PVA associated with this status review expands on this conclusion by using a stochastic individual-based model to evaluate factors (such as disease outbreaks or other chance events) that may threaten small populations, even if these populations on the whole show positive population growth. However, I have two areas of concern with the PVA, and with the resulting conclusion as to the resilience of the current Oregon wolf population:
1) the manner in which stochastic factors are parameterized in the PVA is overly optimistic;
2) the PVA does not incorporate the effects of small population size and isolation on genetic threats to population viability. Instead the status review relies on a brief qualitative discussion which does not accurately represent what is currently known about genetic threats to small wolf populations.
Treatment of stochastic factors The ODFW PVA incorporates stochastic factors such as disease outbreaks or prey decline in two ways (PVA p 14): 1) An effect on reproduction via a 5% chance per pack of reproductive failure in any year. Importantly, these reproductive failures were not correlated between packs, so population-level reproductive output did not experience “bad years”. 2) An effect on population-level survival where survival was reduced by 25% on average once in 100 years.
The PVA does not document the source of these parameter estimates, but they appear highly optimistic when compared to data from well-studied wolf populations such as in the Yellowstone region. In terms of stochastic factors affecting reproduction, effects of disease outbreaks on fecundity (considered broadly to include pup survival) are often correlated between packs in a population, which increase the effect of this factor on viability. Additionally, the ODFW PVA’s mean interval of 100 years between catastrophes likely underestimates the frequency of events impacting population-level survival rates. If
2
WWW.KLAMATHCONSERVATION.ORG Klamath Center for Conservation Research PO Box 104, Orleans, CA 95556 USA
only rare “catastrophic” events are considered, then a 25% decrement likely underestimates the effect of such an event on survival. In contrast to the parameters used in the ODFW PVA, Almberg et al. 2010 concluded based on data for the Yellowstone region that “wolf managers in the region should expect periodic but unpredictable CDV-related population declines as often as every 2–5 years”.
Treatment of genetic issues associated with population size and isolation Recent wolf PVAs (e.g., Carroll et al. 2013) have explicitly incorporated the effects of genetic factors on population viability. In contrast, the ODFW PVA omits quantitative consideration of genetic factors, which may cause its results to be overly optimistic. The status review relies on statements such as “In context of a larger meta-population, Oregon’s wolf population is neither small, nor isolated” (p 20). This statement is so general as to be uninformative. Wolves were historically present throughout their range in the lower 48 states as a largely continuous population with some degree of genetic isolation by distance (Vonholdt et al. 2011). The current Oregon wolf population is small and relatively isolated when compared to historic conditions, and thus genetic factors are of potential concern. This is true even when Oregon’s wolves are considered in a metapopulation context. The fact that wolves are good dispersers even in the current landscape may reduce genetic effects associated with small population size but will not eliminate these effects.
The review implicitly assumes that wolf populations in other states within the metapopulation will remain at their current size and continue to be a robust source of dispersing individuals. For example, on page 18, the document states “We contend that high levels of genetic diversity in Oregon wolves will be maintained through connectivity to the larger NRM wolf population.” However, one cannot assume that populations in adjacent states will remain at current levels. The Idaho wolf population could potentially be reduced fivefold from its recent peak level, to a minimum of 150 wolves, under current state management regulations. Any such reduction would reduce dispersal into Oregon below that evident in the last decade. Additionally, if, in the longer term, hunting is permitted after delisting of Oregon wolves, this increased human-caused mortality, even if sustainable from a demographic perspective, would be expected to reduce immigration from the NRM population.
3
WWW.KLAMATHCONSERVATION.ORG Klamath Center for Conservation Research PO Box 104, Orleans, CA 95556 USA
More generally, the document’s statement (p 17) that “Small populations of wolves are unlikely to be threatened by low genetic diversity” is not consistent with the latest research on small wolf populations. For example, the wolf population in Isle Royale National Park has long been used as an example of the ability of a small, isolated wolf population to persist. However, recent developments have demonstrated the high risks associated with genetic inbreeding in this population (Raikkonen et al. 2009), which as of early 2015 had dwindled to 3 individuals (Vucetich and Peterson 2015). Similarly, the Finnish wolf population has decreased in size in recent years to the point where it has become genetically depauperate (Jansson et al. 2012).
Given these potential risks, a precautionary management approach is appropriate in order to avoid undermining the progress to date in recovering Oregon’s wolf populations. Management of wolves in the Eastern Wolf Management Zone (WMZ) should ensure that the rate of dispersal to western Oregon during the period in which the western population is still being established is not reduced, so that wolf populations in the Western WMZ can be founded with the broadest sample of genetic representation from the larger metapopulation, in order to avoid future genetic problems. Continued frequent dispersal into the Western WMZ will also facilitate the establishment of wolf populations is all “significant portions of range” in western Oregon where habitat remains suitable for wolves.
Sincerely, Carlos Carroll, Klamath Center for Conservation Research, e-mail: [email protected]
Literature Cited Almberg, E.S., P.C. Cross & D.W. Smith. 2010. Modeling the spatial scale and multi-host dynamics of canine distemper virus in Greater Yellowstone Ecosystem carnivores. Ecological Applications 20(7):2058- 2074. Carroll, C., R. J. Fredrickson, and R. C. Lacy. 2013. Developing metapopulation connectivity criteria from genetic and habitat data to recover the endangered Mexican Wolf. Conservation Biology 28:76-86.
4
WWW.KLAMATHCONSERVATION.ORG Klamath Center for Conservation Research PO Box 104, Orleans, CA 95556 USA
Jansson, E., M. Ruokonen, I. Kojola and J. Aspi. 2012. Rise and fall of a wolf population: genetic diversity and structure during recovery, rapid expansion and drastic decline. Molecular Ecology 21(21):5178– 5193. Raikkonen, J., J. A. Vucetich, R. O. Peterson, and M. P. Nelson. 2009. Congenital bone deformities and the inbred wolves (Canis lupus) of Isle Royale. Biological Conservation 142:1025–1031. Vonholdt, B. M., et al. 2011. A genome-wide perspective on the evolutionary history of enigmatic wolf- like canids. Genome Research 21:1294–1305. Vucetich, J.A. and R. O. Peterson. 2015. Ecological Studies of Wolves on Isle Royale: Annual Report 2014–15. Michigan Technological University, Houghton, MI.
5
Appendix C
LIST OF ALL PUBLISHED LITERATURE PROVIDED BY SWG ENVIRONMENTAL CAUCUS TO CDFW ONE YEAR AGO CITATIONS PREVIOUSLY PROVIDED TO CDFW IN JANUARY 2015 (pdf copies of cited materials were supplied to CDFW at that time)
In December 2014, CDFW provided an internal version of the draft wolf plan to SWG members, to review and provide feedback to the Department. SWG members replied with comments/text edits/literature citations in January 2015.
The environmental caucus of the SWG provided 56 citations to published, peer-reviewed research, reports, news articles, legal cases and agency website links, along with our comments and recommended edits. For each citation, we indicated the chapter of the plan in which it should be included and discussed. However, in the public version of the draft wolf plan CDFW released for public comment in December 2015, only one of the 56 citations has been included and discussed.
The 2015 public comment version of the plan is structured slightly differently than the 2014 internal version. The list below is of all 56 citations, indicated by chapter of relevance in the 2015 public version of the plan.
DRAFT WOLF PLAN, PART I
Hayes, R.D. 2010. Wolves of the Yukon. Wolves of the Yukon Publishing. 278 pp.
Gifford Pinchot Task Force v. U.S. Fish & Wildlife Service (9th Cir. 2004) 378 F.3d 1059, 1070
Natural Resources Defense Council v. Fish and Game Com. (1994) 28 Cal.App.4th 1104, 1117-18
THE FOLLOWING CITATIONS ARE FOR CHAPTERS IN DRAFT WOLF PLAN, PART II
CHAPTER 1 – WOLF LIFE HISTORY AND BACKGROUND
Almberg, E.S., P.C. Cross, A.P. Dobson, D.W. Smith, and P.J. Hudson. 2012. Parasite invasion following host reintroduction: a case study of Yellowstone’s wolves. Phil. Trans. R. Soc. B (2012) 367, 2840-2851.
Dumbacher, J., S. Fallon, W. Murdoch, J. Patton, R. Wayne, P. Wilson, and S. Courtney. 2014. Review of Proposed Rule Regarding Status of the Wolf Under the Endangered Species Act. National Center for Ecological Analysis and Synthesis. U.C. Santa Barbara. 68 pp.
Estes, J.A., J. Terborgh, J.S. Brashares, M.E. Power, J. Berger, W.J. Bond, S.R. Carpenter, T.E. Essington, R.D. Holt, J.B.C. Jackson, R.J. Marquis, L. Oksanen, T. Oksanen, R.T. Paine, E.K. Pikitch, W.J. Ripple, S.A. Sandin, M. Scheffer, T.W. Schoener, J.B. Shurin, A.R.E. Sinclair, M.E. Soule, R. Virtanen, and D.A. Wardle. 2011. Trophic downgrading of Planet Earth. Science 333: 301.
McIntyre, R. 1995. War Against the Wolf. America’s Campaign to Exterminate the Wolf. Voyageur Press, Inc. Stillwater, MN. 495 pp., at pp. 120-121.
Newland, M. and M. Stoyka. 2013. The Pre-contact distribution of Canis lupus in California: A preliminary assessment. Anthropological Studies Center, Sonoma State University. Rohnert Park, California. 20 pp.
Peterson, R.O., Vucetich, J.A., Bump, J.M. and D.W. Smith. 2014. Trophic Cascades in a Multicausal World: Isle Royale and Yellowstone. Annual Review of Ecology, Evolution and Systematics. 45: 325-45. Doi 10.1146/annurev-ecolsys-120213-091634.
Ripple, W.J., J.A. Estes, R.L. Beschta, C.C. Wilmers, E.G. Ritchie, M. Hebblewhite, J. Berger, B. Elmhagen, M. Letnic, M.P. Nelson, O.J. Schmitz, D.W. Smith, A.D. Wallach, and A.J. Wirsing. 2014. Status and ecological effects of the world’s largest carnivores. Science Vol. 343: 1241484.
Rutledge, L.Y., Patterson, B.R., Mills, K.J., Loveless, K.M., Murray, D.L., and B.N. White. 2010. Protection from harvesting restores the natural social structure of eastern wolf packs. Biological Conservation 143: (2010) 332-339.
Seton, E.T. 1929. Lives of Game Animals, Vol 1: Cats, wolves, and foxes. Doubleday and Doran, Co., New York.
USFWS, Arizona Game and Fish Department, USDA-APHIS Wildlife Services, US Forest Service, and White Mountain Apache Tribe. 2014. Mexican Wolf Recovery Program 2013 Interagency Annual Report.
Weiss, A., Greenwald, N., and C. Bradley. 2014. Making Room for Wolf Recovery. The Case for Maintaining Endangered Species Act Protections for America’s Wolves. Center for Biological Diversity. 24 pp.
HSUS vs. Sally Jewel et al. U.S. District Court for the District of Columbia Memorandum Opinion, issued December 19 2014.
CHAPTER 2 – DISEASES AND WOLVES None
CHAPTER 3 – HUMAN INTERACTIONS AND CURRENT PERCEPTIONS OF WOLVES
Blumm, M.C. and A. Paulsen. 2013. The Public Trust in Wildlife. Utah Law Review, No. 6: 1437- 1504.
Chambers, Catherine M., and John C. Whitehead. 2003. A contingent valuation estimate of the benefits of wolves in Minnesota. Environmental and Resource Economics 26:249-67.
Davis and Hibbitts, Inc. 1999. Telephone poll of 600 registered Oregon voters, focused on possible return of wolves to Oregon, conducted April 6-8, 1999. Portland, Oregon: Davis and Hibbits, Inc.
Duffield, John W. 1992. An economic analysis of wolf recovery in Yellowstone: Park visitor attitudes and values. In Wolves for Yellowstone? A report to the United States Congress. Vol. 4, research and analysis, eds. John D. Varley, and Wayne G. Brewster, 2-87. Yellowstone National Park, Wyoming: National Park Service, Yellowstone National Park.
Duffield, John W., and Chris J. Neher. 1996. Economics of wolf recovery in Yellowstone National Park. In Transactions of the 6Jst North American wildlife and natural resources conference, eds. Kelly G. Wadsworth, and Richard E. McCabe, 285-92. Washington, DC: Wildlife Management Institute.
Duffield, Jon. W., D. A. Patterson, and C. J. Neher. 2008. Wolf recovery in Yellowstone Park visitor attitudes, expenditures, and economic impacts. Yellowstone Science. 16(1):20-5.
Krutilla, John V. 1967. Conservation reconsidered. American Economic Review. 56:777- 86.
Lee, Y., Harrison, J.L., Eisenberg, C. and B. Lee. 2012. Modeling biodiversity benefits and external costs from a keystone predator reintroduction policy. Journal of Mt. Science 9:385- 394. Doi10.1007/s11629-009-2246-1.
Manfredo, M. J., A. D. Bright, J. Pate, and G. Tischbein. 1994. Colorado residents' attitudes and perceptions toward reintroduction of the gray wolf (Canis lupus) into Colorado. Fort Collins, Colorado: Colorado State University, Human Dimensions in Natural Resources Unit.
Rosen, W. 1997. Recovery of the red wolf in northeastern North Carolina and the Great Smoky Mountains National Park: An analysis of the social and economic impacts, report submitted to the U.S. Fish and Wildlife Service and the Point Defiance Zoo and Aquarium, 31- 6. Wolves of America Conference Proceeding.
Schoellhamer, M. 2013. Estimating the economic benefits of a Mt St Helens wolf population: a cost benefit analysis using benefit transfer. Masters Thesis, School of Environmental and Forest Sciences, University of Washington.
Tulchin Research. 2013. Telephone survey of registered voters in California, Oregon and Washington. Tulchin Research Polling and Strategic Consulting.
Weiss, A.E., Kroeger, T., Haney, J.C. and N. Fascione. 2007. Social and ecological benefits of restored wolf populations. Transactions of the 72nd North American Wildlife and Natural Resources Conference. Pp. 297-319.
Center for Biological Diversity v. FLP Group, Inc. 83 Cal.Rptr.3d 588 (Cal. Ct. App. 2008).
CHAPTER 4 – WOLF AND DOMESTIC DOG INTERACTIONS
Patronek, G.J., Sacks, J.J., Delise, K.M., Cleary, D.V., and A.R. Marder. 2013. Co-occurrence of potentially preventable factors in 256 dog bite-related fatalities in the United States (2000- 2009). J Am Vet Med Assoc. 2013 Dec 15;243(12):1726-36. doi: 10.2460/javma.243.12.1726. Sacks, J.J., MD, MPH; Sinclair, L., DVM; and J. Gilchrist, MD (15 September 2000). "Special Report: Breeds of dogs involved in fatal human attacksin the United States between 1979 and 1998". JAVMA, Vol 217, No. 6. JAVMA. Retrieved 2013-06-11.
Stark, Mike. 2006. UM economist: Wolves a big moneymaker. Billings Gazette. http://www. billings gazette .net/articles/2006/04/07 /news/state/25- wolves.txt.
CHAPTER 5 – WOLF INTERACTIONS WITH OTHER WILDLIFE SPECIES Wilmers, C.C. and W.M. Getz. 2005. Gray wolves as climate change buffers in Yellowstone. PLoS Biology. 3(4):e92.
CHAPTER 6 – WOLF INTERACTIONS WITH UNGULATES Hobbs, N. Thompson. 4/12/2006. A Model Analysis of Effects of Wolf Predation on Prevalence of Chronic Wasting Disease in Elk Populations of Rocky Mountain National Park. Unpublished report for Rocky Mountain National Park.
Montana Fish, Wildlife and Parks, and University of Montana. 2014. Bitterroot Elk Project Progress Report. Spring 2014. 4 pp.
Webb, 2014. Results of environmental scanning applied to the design of a deer management design support system (DSS) for the United States and California. Issues in Information Systems. Vol. 15 (II): 77-88. San Jose State University.
Wild, M.A., Hobbs, N.T., Graham, M.S. and M.W. Miller. 2011. “The role of predation in disease control: A comparison of selective and non-selective removal of prion diseases in deer.” Journal of Wildlife Diseases 47(1):78-93. CA Essential Habitat Connectivity Report https://www.wildlife.ca.gov/Conservation/Planning/Connectivity https://www.wildlife.ca.gov/Conservation/Planning/Connectivity/CEHC
National Fish, Wildlife and Plant Climate Adaption Strategy http://www.wildlifeadaptationstrategy.gov
Montana Fish, Wildlife & Parks website reports on elk population, distribution and objectives, at http://fwp.mt.gov/fishAndWildlife/management/elk/
WY elk at 34% over objective (2010) http://www.jhnewsandguide.com/news/top_stories/gyc-sees-wolf-solution/article_ae0425ba- 380c-5c2a-90dc-067e41c6d026.html
2011 WY elk harvest success http://wgfd.wyo.gov/web2011/news-1001370.aspx
2013 WY elk harvest success http://wgfd.wyo.gov/web2011/news-1001959.aspx;
CHAPTER 7 – EFFECTS OF WOLVES ON LIVESTOCK AND HERDING/GUARDING DOGS
Wielgus, R.B. and K.A. Peebles. 2014. Effect of wolf mortality on livestock depredations. PLoS ONE 9(12): e113505. Doi:10.1371/journal.pone.0113505
CHAPTER 8 – COORDINATION WITH OTHER STATES AND FEDERAL AGENCIES
US Department of Agriculture. 2012. National Forest System Land, Management Planning. Federal Register Vol. 77, No. 68. Accessed January, 2015. http://www.fs.usda.gov/Internet/FSE_DOCUMENTS/stelprdb5362536.pdf.
CHAPTER 9 – WOLF CONSERVATION
Bryan, H.M., Smits, J.E.G., Koren, L., Paquet, P.C., Wynne-Edwards, K.E., and M. Musiani. 2014. Heavily hunted wolves have higher stress and reproductive hormones than wolves with lower hunting pressure. Functional Ecology. doi: 1111/1365-2435.12354.
Hedrick, P. W. and R. Fredrickson. 2010. Genetic rescue guidelines with examples from Mexican wolves and Florida panthers. Conserv Genet 11:615–626
Soule, M.E., Estes, J.A., Miller, B. and D.L. Honnold. 2005. Strongly interacting species: Conservation policy, management, and ethics. Bioscience 55(2): 168-176.
Soule, M.E., Estes, J.A., Berger, J. and Martinez Del Rio, C. 2003. Ecological effectiveness: conservation goals for interactive species. Conservation Biology 17(5): 1238-1250. Oct 2003.
Vucetich, J.A. and R.O. Peterson. 2014. Ecological studies of wolves on Isle Royale. Annual Report 2013-2014. School of Forest Resources and Environmental Science, Michigan Technological University, Houghton, Michigan USA. 23 April 2014.
Wayne, R. and P Hedrick. 2011. Genetics and wolf conservation in the American West: lessons and challenges. Heredity 107:16-19.
Ace II Mapping http://www.dfg.ca.gov/biogeodata/ace/
Department’s 2009 Climate Plan for Habitat and Biodiversity http://www.climatechange.ca.gov/adaptation/documents/Statewide_Adaptation_Strategy_- _Chapter_5_-_Biodiversity_and_Habitat.pdf
Hunter kills wolf in Utah, saying he thought it was a coyote http://www.sltrib.com/news/1999741-155/utah-hunter-kills-wolf-near-beaver
CHAPTER 10 – PUBLIC INFORMATION AND OUTREACH None
CHAPTER 11 – FUNDING NEEDS AND OPPORTUNITIES None
CHAPTER 12 – PLAN EVALUATION AND REPORTING None
Appendix D
WHITE PAPER FROM UNIVERSITY OF WASHINGTON OCTOBER 29, 2014 SYMPOSIUM ON ECOLOGICAL AND SOCIAL COMPLEXITIES OF LETHAL CONTROL OF WOLVES Tackling Wolf Management’s Thorniest Issue: The Ecological and Social Complexities of Lethal Control
A Wolf Conservation Discussion Panel, co-hosted by Pacific Wolf Coalition & University of Washington’s School of Environmental and Forest Sciences (October 29th, 2014)
Catherine H. Gowan1, Carol Bogezi2, John M. Marzluff3, and Aaron J. Wirsing4
School of Environmental and Forest Sciences, College of the Environment, University of Washington, Box 3521000, Seattle, WA 98195-2100, USA
The presence of gray wolves (Canis lupus) evokes intense emotions and attitudes throughout the species’ range (Lute et al., 2014, Treves and Karanth, 2003). Therefore, the return of the gray wolf to the Pacific Northwest has thrilled conservationists but has at times been challenging for some residents, especially hunters and livestock producers, who share turf with these carnivores. In Washington State, wolf recolonization has heightened urban-rural divisions and management efforts to satisfy diverse stakeholders, including conservation groups, livestock producers, and hunters have been complicated. Between 2008, when the first wolf pack was established in Washington, and 2015, two packs in Washington have repeatedly depredated livestock and in response have been targeted with lethal control. In 2012, the Wedge Pack was removed for depredating cattle and in 2014, the breeding female of the Huckleberry Pack was removed after the pack depredated 24 sheep. Both uses of lethal control were controversial and led to extensive public comment, highlighting the need for scientific data to inform the use of lethal control as a management tool.
Understanding the social acceptability of the presence and management of predators is vital to their conservation in a human-dominated world. Although approximately 64% of Washington
1 Email: [email protected]
2 Email: [email protected]
3 Email: [email protected]
4 Email: [email protected] State residents are in favor of wolf recovery (Responsive Management, 2014), the wolf debate remains intense and management decisions by the Washington Department of Fish and Wildlife (WDFW) are under heavy scrutiny. Approximately 63% of Washington residents support some level of lethal removal to protect livestock, while 28% are opposed (Responsive Management, 2014). Residents’ values concerning wolves often determine their attitudes toward wolf recovery and the management of wolves in the State. In a survey of Washington residents’ attitudes towards wolves, respondents whose views of wildlife were either focused solely on utilization or open to multiple views of nature and conservation were less accepting of wolf recovery than other value orientation types (Dietsch et al., 2011). In general, residents living in the eastern half of the State were less supportive of wolf recovery and more in favor of lethal control of wolves, whereas most residents in the western half of the state were in favor of wolf recovery and less in favor of lethal control of wolves. These divergent values and management priorities have made wolf management in Washington a contentious topic. WDFW is consequently in the position of resolving wolf conflicts that are based as much, if not more, on social values than the biological reality of wolves.
Conflicts over the continued presence of wolves can, at times, be aggravated more by human values, perceptions, and attitudes towards wolves than by economic losses stemming from wolf depredations (Treves and Bruskotter, 2014; Dickman, 2010). At the same time, there is growing concern that management actions aimed at reducing the impacts of predators like wolves on humans interests (e.g., predation on livestock and wild game) could disrupt the social behavior and/or alter the ecological role of these top predators (Wallach et al., 2009; Ordiz et al., 2013). Thus, reconciling the interests of wolf conflict mitigation and conservation requires understanding the social dynamics of both wolves and humans. Fortunately, Washington State is in the early stages of wolf recolonization efforts and can draw upon a growing body of research on the ecology and behaviors of wolves in the wild as well as human dimensions of wolf recovery conducted in other states to inform its own management strategy.
To this end, on October 29th, 2014, University of Washington Professors John Marzluff and Aaron Wirsing and the Pacific Wolf Coalition (pacificwolves.org) hosted a panel discussion of scientists researching issues surrounding one of wolf management’s most controversial aspects –
2 lethal control of wolves. The purpose of this panel discussion was to understand some of the complexities of lethal wolf removal, both social and ecological, in order to inform Washington’s wolf management policies with the best natural and social science available. This effort involved panelists from a wide range of disciplines and life experiences where wolves and people have had extensive interactions. The panelists included: Dr. Scott Brainerd, from the Alaska Department of Fish and Wildlife (ADFG); Dr. Douglas Smith, from Yellowstone National Park; Dr. Robert Wielgus, from Washington State University; Dr. Jeremy Bruskotter, from Ohio State University; and Dr. Adrian Treves, from the University of Wisconsin-Madison.
Dr. Scott Brainerd is the Research Coordinator for the Interior and Northeastern Arctic Region of the Division of Wildlife Conservation with ADFG in Fairbanks, Alaska. He has done extensive research on the impacts of breeder loss on wolf pack social structure in Alaska and Scandinavia. His studies highlight the importance of breeding wolves in maintaining group unity at the pack level.
Dr. Douglas Smith is a senior wildlife biologist for the National Park Service in Yellowstone National Park, Wyoming, and has studied wolf biology for more than 30 years. He has co- authored multiple papers studying how human-induced mortality of individual wolves affects wolf social dynamics and connectivity.
Dr. Robert Wielgus is an associate professor and the director of the Large Carnivore Conservation Lab at Washington State University in Pullman, Washington. He has done extensive research on the impacts of hunting and lethal control on cougars in Washington State. Most recently, he has been studying the effects of lethal control of wolves as a means of reducing livestock depredations.
Dr. Jeremy Bruskotter is an associate professor in the School of Environment and Natural Resources at Ohio State University in Columbus, Ohio. His research areas include natural resources and recreation conflicts and the use of psychology and communication theories in natural resource management and policy.
3
Dr. Adrian Treves is an associate professor at the University of Wisconsin–Madison in Wisconsin. His research has focused primarily on public attitudes toward wolves and wolf policy in Wisconsin, behavioral ecology of carnivores and the risks for people living near them, and methods for mitigating human-wildlife conflicts.
Overview of Wolves in the Pacific Northwest
The purpose of the panel was to inform wolf management actions in Washington State. Thus, it began with an up-to-date overview of wolf management in Washington and the surrounding recovery area. Dr. Donny Martorello, from the WDFW, provided some local context, covering current wolf policy in Washington and Oregon, and Carter Niemeyer, retired U.S. Fish and Wildlife Service (USFWS), covered the broader northern Rockies recovery area and the current wolf population and harvest numbers in Montana, Idaho, and Wyoming. At the time this panel was held, Dr. Martorello was the Carnivore Section Program Manager at the WDFW and he has been at the forefront of assessing the status and management of wolves in Washington since their return to the state in 2008. He currently holds the title of Wolf Policy Lead and oversees much of the ecological recovery and management of wolves in Washington.
Dr. Martorello began the session by giving a brief history of what the WDFW has been doing since wolves arrived in Washington State. In 2007, prompted by the eminent likelihood of wolves recolonizing the state in the next few years, the WDFW Director appointed a 17-member citizen advisory group (the Wolf Working Group or WWG) to advise the development of a state wolf management plan. In addition to the WWG, the WDFW received almost 65,000 public comments on the draft plan and held 23 public meetings around the State for input into the wolf plan. The plan went through a rigorous scientific peer review process. The WDFW Commission approved the Wolf Management and Conservation Plan in 2011 (from here forward referred to as the Wolf Plan). The Wolf Plan is the policy document used to guide recovery and management of wolves in Washington State. Components of the Wolf Plan that involve lethal control of wolves are not implemented in the western two-thirds of the state, since wolves currently remain federally-listed there.
4
In 2013, the WDFW created the Wolf Advisory Group (WAG) to help inform management and guide implementation of the Wolf Plan. The WAG has been asked to review and recommend conflict-reducing strategies and they have also been tasked with being the review board for livestock compensation programs. It is comprised of a diverse group of stakeholders appointed by the WDFW Director. For 2013-2014, there were nine members representing: Quad-county commissioners, the Farm Bureau, Washington Cattlemen’s Association, Cattle Producers of Washington, Conservation Northwest, Humane Society of the United States, Wolf Haven 5 International, Sierra Club, and Hunter’s Heritage. The population trend of wolves in Washington is on the rise, going from one confirmed pack in 2008 to 16 confirmed packs in 2014. The WDFW has been capturing and instrumenting wolves with GPS collars to enable monitoring and assessment of their recovery progress. Washington is seeing a pattern similar to that observed in many Northern Rocky Mountain (NRM) States; namely, relatively high wolf population growth rates, with less than 20% of the wolf packs depredating livestock. Two depredation events, mentioned previously, have occurred in northeastern Washington, resulting in the removal of eight total wolves. In both incidents, the WDFW implemented lethal removal of the problem wolves, in accordance with the Wolf Plan (page 80).6 The WDFW has a checklist for non-lethal preventive measures to be followed before lethal control efforts are considered. The checklist is composed of the following five preventive non-lethal tools that livestock owners are advised to implement before lethal control is used: removing livestock carcasses, removing sick and/or injured livestock, securing bone yards, calving or lambing away from wolves, and hazing wolves if they are encountered. These preventive non-lethal tools are required before WDFW implements any lethal control action on wolves, but it is only recommended that they be in place before depredations occur (Appendix, Figure 1).
In 2008, WDFW began allocating resources toward hiring conflict-specialists, employees acting as liaisons between the WDFW and landowners on wildlife conflict issues, for every region, setting up a compensation program for livestock loss due to wolves, and entering into cooperative, cost-share agreements with livestock owners to implement preventative measures
5 In the fall of 2014, WDFW decided to expand the membership of the WAG and, as of 2015, there are now 17 individuals on this advisory body, including representatives from additional organizations.
6 Conservation groups disagreed with WDFW actions in both lethal removal instances, citing literature contained in the Wolf Plan (81).
5
(known as depredation prevention cooperative agreements). Thus far there are 137 conflict specialists (Appendix, Figure 2) working with livestock owners and the general public in areas with the highest level of wildlife conflict (either predator or elk and deer related conflict). There are currently 41 active damage prevention cooperative agreements in the state. These agreements can provide funds for improved fencing, sanitation, guard animals, range riders, and other preventative measures.
Dr. Martorello also presented a snapshot of the Oregon Wolf Management Plan on behalf of Russ Morgan, the Oregon Department of Fish and Wildlife (ODFW) Wolf Program Coordinator. The ODFW developed a wolf conservation and management plan in 2005 and updated it in 2010. As in Washington, there was extensive public input on the wolf management plan, and a Wolf Advisory Group consisting of 14 members who represent its primary stakeholder groups was assembled to advise ODFW during the creation of the plan. The recovery trends are similar to Washington State and the NRM States. With 10 packs and about 64 wolves in Oregon as of October 2014, ODFW expects to reach their recovery objective of 4 successful breeding pairs for three consecutive years by 2015.8 As of 2013, there was an addendum to the wolf plan with new rules on the use of lethal control and harassment of wolves: the new rules set the bar at four qualified depredations by the same pack over the course of six months before lethal control can be considered. Livestock owners must not have unnatural attractants and must implement one non-lethal measure before a depredation will be considered qualified. It is mandatory for ranchers to implement at least one non-lethal wolf control measure as part of their animal husbandry (including removal of any possible wolf attractants including dead animals and any calving afterbirth remains) at least seven days prior to and on the day of the depredation (ODFW Wolf Plan, 2010). In Oregon, when a depredation is suspected, the livestock operator must then implement at least one non-lethal site-specific measure before any additional depredations are considered qualified for potential lethal action against the wolf pack. If the livestock operator implements appropriate non-lethal measures for a period of six months and three depredations by the same wolf or wolf pack occur, ODFW will assess whether the offending wolf or wolf pack is likely to continue depredating, regardless of increased implementation of non-lethal measures.
7 At the date of this publication, there are 19 conflict specialists, 11 of whom work with livestock producers in areas with wolf packs. 8 ODFW’s 2014 wolf count observed 77 individuals.
6
After this assessment lethal control may be approved. In addition to these new rules, the ODFW is making an effort to be as transparent as possible, using the agency website (http://www.dfw.state.or.us/wolves/) to post information about preventative measures, depredation investigations, and pack locations and timelines9.
Wolf recovery in the Northern Rocky Mountain (NRM) recovery areas
Carter Niemeyer has been a primary player in wolf recovery since wolves were first reintroduced in Yellowstone. He is retired from the USDA Wildlife Services and US Fish & Wildlife Service (USFWS) in Idaho where he worked from the mid-1980s until 2006. Mr. Niemeyer was part of the team that captured and reintroduced the wolves in Yellowstone National Park and Idaho in 1995/1996. He gave a summary of events in the broader recovery area and the wolf population trends and harvest numbers in Montana, Idaho, and Wyoming from 1982 to 2013. Recovery in the NRM was achieved in 2002, and currently there are at least 1,657 wolves and 282 packs. In most NRM states wolf populations have remained stable and are federally delisted, with the exception of Wyoming whose limited state regulation to protect wolves outside of Yellowstone National Park has resulted in re-listing their wolf population as endangered. Mr. Niemeyer emphasized that the use of non-lethal management tools is not mandatory in most States, Washington included, except for the consideration of lethal control. Consequently, he found that it is not readily implemented by many producers. However, by all biological measures the NRM wolf population remains secure under state management and he called gray wolf reintroduction “an amazing success story.”
Summary In summary, numbers of wolves and breeding pairs in Oregon and Washington are increasing steadily, promising the real possibility of achieving the two states’ recovery goals by 2015 and 2021, respectively. Challenges remain in the continued use of the best available science to solve human-wolf conflicts in Washington, as in other western states where wolves are recovering.
9 The Oregon Wolf Plan has a three-phased approach and the new rules described here apply only during Phase I of the Wolf Plan. Once wolves in Oregon reach population and breeding pair thresholds that move them out of Phase I and into Phase II, these new rules no longer apply.
7
This discussion panel served to address the difficulties highlighted by Dr. Martorello and Mr. Niemeyer.
Review of Current Research by the Panelists
Dr. Scott Brainerd
Dr. Scott Brainerd based his talk on two recent articles concerning the effects of breeder loss on wolf populations (Brainerd et al., 2008; Borg et al., 2015). The first article, a meta-analysis of 148 breeding wolves in both national parks and unprotected areas, showed that more than half of breeder loss was from anthropogenic causes. The authors found that pups survived in 84% of the cases where breeders were lost, regardless of the sex of the breeder, and that pup survival was better in larger packs (greater than six wolves) where the pups were over six months of age. The loss of both breeders was far more detrimental to reproduction than just the loss of one; 56% of packs that only lost one breeder reproduced subsequently whereas only 9% of packs that lost both breeders subsequently reproduced. Also, relatively few of the packs dissolved after losing one breeder (38.2%), while a much larger proportion dissolved after losing both breeders (>80%). In areas where wolf populations were saturated, it took less time for packs to replace breeders (10 months) than in areas where wolves were recolonizing (~20 months).
Borg et al. (2015) based their article on a 36-year dataset on wolves in Denali National Park where much of the breeder loss was from natural causes, primarily from other wolves. In Denali, breeder loss accounted for 77% of pack dissolution. Packs were more likely to dissolve if breeder loss was due to anthropogenic causes than natural ones. However, there was no demonstrable effect of breeder loss on the overall population of wolves in Denali.
Dr. Robert Wielgus
Dr. Robert Wielgus presented the results of his research assessing the efficacy of wolf lethal control in preventing livestock depredations using data collected from 1987 to 2012 in Idaho, Montana, and Wyoming (Wielgus and Peebles, 2014). He highlighted the unexpected relationship between lethal control of wolves and livestock depredation rates. Specifically, his team used the annual USFWS wolf harvest reports from 1987-2012 and the United States Department of Agriculture National Agricultural Statistics Service (USDA NASS) records of
8 cattle and sheep depredation counts from wolf occupied counties to determine the correlation between lethal control and livestock depredation incidents. The results of the analysis showed that an increase in livestock, breeding pairs, and wolf numbers was correlated with increased instances of depredation. However, countering common wisdom that wolf removal decreases livestock depredations, Wielgus and Peebles found that there was a 5% increase in depredation (both for cattle and sheep) for every wolf that was killed. This phenomenon is possibly caused by an increase in breeding pairs to compensate for lethal wolf removal (Wielgus and Peebles, 2014). Ultimately, it seems that lethal control might decrease depredation on a local scale, but may not control widespread livestock loss unless over 25% of the wolf population is lethally removed.
Dr. Douglas Smith
Dr. Douglas Smith began his talk by emphasizing that wolves in Yellowstone National Park allow us to study the ecology of these carnivores in an environment mostly free of human interference. Since the reintroduction of wolves into the park in 1995 and 1996, Dr. Smith has been studying pack size and complexity to determine how unexploited packs behave in the wild. Dr. Smith has found that, in the absence of management, wolf packs quickly become socially complex and retain multiple age groups. Having a variety of age groups in a pack is important for the segregation of hunting duties, although not necessarily hunting success (MacNulty et al., 2012; Mech, 1999); females and younger males are faster, while older males are bigger and stronger (MacNulty et al., 2009). In contrast, many of the wolf packs outside of Yellowstone protected area are simple packs composed of breeders and offspring (Smith, unpublished data). Wolf packs are generally composed of primarily younger wolves even in protected areas due to intraspecific strife (Mech, 1994). However, more complex packs are more likely to survive territorial disputes and hunt efficiently than simple packs (Smith, unpublished data). Smith and his colleagues are also finding that pups act as social glue for wolf packs; the more pups born to a pack, the less likely it is that subordinate wolves will disperse (Smith, unpublished data). In addition, dispersal patterns are seasonally dependent inside Yellowstone National Park, whereas outside the park there is no observable seasonal pattern (Smith, unpublished data). At this time, more research is needed to explain why this disparity might exist.
Dr. Smith concluded that it appears that killing wolves reduces social cohesion within a pack. For example, removal of the breeder female was shown to result in higher reproductive rates in a
9 pack due to the breeder male mating with more than one remaining female in the pack. Killing of wolf pups can cause stress in the family pack, and packs whose pups are killed are more likely to disperse than to stay together. Dispersing wolves are more likely to cause livestock conflicts as they find new geographic areas to occupy. In addition, killing wolves may reduce the packs to sizes that cannot efficiently hunt game species and resort to easy prey such as sheep.
Dr. Adrian Treves
Transitioning from the ecological to the social sciences, Dr. Adrian Treves presented his research on common assumptions that lead to lethal removal of wolves. Generally, there are two main justifications for lethal removal of wolves: prevention of property damage and improvement of attitudes towards carnivores (Treves et al., 2009). Dr. Treves believes both of these assumptions need to be evaluated, especially in light of the recent finding by Wielgus and Peebles (2014) that lethal management of wolves is correlated with increased frequency of livestock depredations. One of the ways Dr. Treves studied this issue in Wisconsin was by developing a risk map to determine which areas were most likely to have wolf-livestock conflict. By examining differences between sites with and without depredations, Dr. Treves determined that areas with more grassland and pasture, closer to known wolf pack ranges, and farther from forest coverage were at the highest risk for depredations (Treves et al., 2011). This risk map proved over 90% effective at predicting depredation sites. A highly predictive risk map allows managers to anticipate and plan for, rather than react to, conflicts. Additionally, risk maps can be used to guide local action and reduce the perceived risk of wolf conflicts among residents living close to wolf territories. Dr. Treves found that lethal removal has mixed results in terms of protecting private property. Namely, he found that the time period between recurring depredations is actually shortened after lethal removal than after use of non-lethal control measures in Michigan (Treves, 2013).
Thus far, lethal removal has also not been found to improve attitudes toward wolves. Treves conducted two surveys of two different citizen panels. The first panel from 2001 was chosen for wolf experience and hunting (Naughton-Treves et al., 2003), the second from 2004 was a random sample (Treves et al., 2009) and both were resampled in 2009. In the years between the surveys, wolf numbers and depredations increased and lethal control of wolves was increased. There was a significant amount of media attention centered on wolf issues. Instead of showing a
10 positive trend in attitudes with increased use of lethal control, a significant portion of citizens became more negative about wolf populations; 37% of respondents reported an increase in the likelihood they would shoot a wolf if they saw it, 44% reported an increased agreement with the statement that Wisconsin’s wolf population threatened deer hunting opportunities, and 46-47% of the respondents showed increased agreement to hunting wolves (Treves et al., 2013). On the other hand, support for government-sponsored lethal removal increased. Attitudes continued to decline after one year of a public hunting and trapping season (Hogberg et al., 2013). It appears that legalizing the killing of wolves devalued the wolf in the eye of the public in Wisconsin (Treves and Bruskotter, 2014).
Dr. Jeremy Bruskotter
Even if attitudes towards wolves on a local scale are becoming more negative in areas where wolves are present, attitudes towards carnivores and other traditionally maligned animals are becoming more positive on a national scale. Dr. Jeremy Bruskotter presented his work on the psychology underlying reactions to large carnivores (Bruskotter et al., 2007; Bruskotter, 2011; Slagle et al., 2013). Dr. Bruskotter (2011) replicated a study by Kellert (1978) surveying the general public about their feelings toward 26 different species of animals. The results showed a marked increase in positive feelings towards wolves; there was a 9% increase of participants indicating they felt “very positively” about wolves and a 3% decrease in respondents who felt “very negatively” about wolves. It is important to study the mental processes, such as perceived risks and benefits and emotional responses, which go into our cognitive construction of animals like wolves because they affect both policy and behavior, potentially leading to intolerance or coexistence. With this in mind, Dr. Bruskotter surveyed readers of an active wildlife blog, www.thewildlifenews.com, about their perceived risks and benefits of wolves and their affective responses to wolves. He found that perceived benefits, such as healthier riparian areas, were more predictive of support of wolves than perceived risks, such as dangers to livestock and children, and that there was a large indirect effect of emotional reactions to wolves on perceived benefits (Slagle et al., 2012). In other words, positive feelings towards wolves strongly affect the belief in perceived benefits and subsequent tolerance for wolves. However, negative emotional responses lead directly to intolerance towards wolves, rather than simply an increase in perceived risks (Treves and Bruskotter, 2014). Dr. Bruskotter posited that an individual’s perception of
11 wolves is rooted in group identity and membership. Therefore, social conflict must be taken into account when managing wolves. For example, the residents in the NRM can be split into the “new west” (animal rights advocates, environmentalists, conservationists, wildlife advocates) and the “old west” (hunters, gun rights advocates, farmers/ranchers, property rights advocates), and this delineation predicts beliefs about positive impacts of wolves (Bruskotter, unpublished data). Because individual perceptions of wolves can be influenced by group membership, elite cues-- such as a message from an influential member of the community-- can have enormous impact on the individuals in the group and shape how they view wolves as well as other groups. In terms of management implications, it is important to focus on shared goals and improvements (rather than just solutions to problems), build trust through cooperative efforts and avoid “demonizing” the other side, and try to avoid power structures that favor some groups over others (Bruskotter, 2014).
Implications for Washington State
The research findings presented by the panelists can be incorporated into management decisions and help inform future wolf management and conservation strategies in Washington State. The following section will cover the take-away points of the discussion panel and what they mean for Washington State going forward. However, before these specifics, we wish to point out a more general suggestion. The wolf research community is large and scientifically-focused. As such, the WAG would benefit from regular consultation with outside scientists, and might consider expanding its membership to include at least one wolf researcher.
A common scientific finding is that lethal management of wolves has many unintended consequences on wolves and human perceptions of wolves. Although Dr. Brainerd’s research showed the remarkable resilience of wolves faced with breeder loss, there are still negative effects on wolf packs, such as simplification of social structure, pack dissolution and short-term reproduction decreases, especially when wolves are newly recolonizing an area (Brainerd et al., 2008; Borg et al., 2015). Furthermore, removal of wolves may increase future livestock depredation and has been found to lower the public’s valuation of a wolf’s life (Wielgus and
12
Peebles, 2014; Treves and Bruskotter, 2014). These findings suggest five lessons Washington managers can apply to minimize human-wolf conflict.
1. Consider the needs of a recolonizing population.
One aspect of Dr. Brainerd’s work that is highly relevant for wolf management in Washington is the finding that wolf packs in areas that have already been recolonized and saturated replaced their lost breeders more quickly than wolf packs in recolonizing areas. By implication, wolf packs in areas that have high connectivity with other wolf packs may be more resilient to breeder loss than wolf packs in recolonizing areas (Brainerd et al., 2008, Borg et al., 2015). It is possible, in light of Dr. Brainerd’s work, that the shooting of the Huckleberry Pack female breeder in August 2014 will cause the Huckleberry pack to dissolve into solitary sub-adult wolves seeking their own territories10. However, the Huckleberry pack’s proximity to the wolves in NRM recovery area and Canada will likely diminish the effects of breeder loss because new females can be recruited from nearby packs fairly rapidly. More isolated packs in the central portion of the state, such the Teanaway or Lookout packs, will need to be managed more carefully. Breeder loss may increase the chance of wolves coming into conflict with livestock due to two highlighted mechanisms: (1) either inexperienced young wolves lacking the pack complexity to hunt large ungulates such as elk (Smith, unpublished) will prey on sheep, or (2) as a result of compensatory reproduction due to non-breeding females breeding in the absence of the breeder female (Borg et al., 2015). Therefore, unintended consequences of lethal control could include a delay in achieving wolf recovery goals and an increase in livestock conflicts. For example, in late October of 2014 the breeding female of the Teanaway pack, which had not depredated any livestock, was illegally poached. Later in the summer of 2015, the pack was attributed with having depredated a calf, possibly as an unintended consequence of killing the alpha female. Wielgus and Peebles (2014) found that 25% of wolves must be removed to decrease livestock depredation. For wolves in the recovery phase, this percentage is high and if implemented the recovery goals will take longer to achieve and wolves will continue to be federally listed.
10 In August of 2015, Dr. Martorello briefed WAG members on the status of the Huckleberry pack and noted that it is the WDFW’s belief that the Huckleberry pack has split into two packs – a northern pack which contains the original breeding male and several other wolves, and a southern pack, which contains remaining members of the original pack.
13
2. Provide wild space for wolves.
Washington wolf managers may want to consider, as Dr. Smith suggested, what it means to manage for “naturalness” in wolf populations. Between the effects of breeder loss, the tendencies of wolves to form complex rather than simple packs, the importance of key individuals for hunting success, and the unexpected relationship between increased lethal removal and increased livestock depredations (Brainerd et al., 2008; MacNulty et al., 2011; Wielgus and Peebles, 2014), the science is painting a complicated picture for management agencies using lethal control. Given the complexities of managing wolves with lethal control and the risks of such control during the early phases of recolonization (Brainerd et al., 2008; Borg et al., 2015), it would be reasonable to first manage for recovery by allowing wolves sufficient wild space. Washington State is highly populated, however, and while there are wilderness areas that can provide habitat for wolves, they are not as extensive as those in other western states. Consequently, Washington may benefit from a zonal approach that expands on refuges for wolves with management that mimics wild space. These strategic areas could provide extensive protection for wolves, even after delisting, while other areas would allow for more active wolf management (lethal control and/or regulated hunting).
3. Develop a predictive map of the risk of human-wolf conflict.
A risk map (see Treves et al. 2011 for an example) of Washington’s wolves showing which ranchers are most likely to experience depredations would be helpful to streamline efforts to work proactively with ranchers. Most ranches, even those with wolf packs in their proximity, do not experience depredations often and as such may not be motivated to enroll in WDFW cooperative agreements or implement non-lethal measures. Livestock producers and WDFW officials could use a risk map to assess whether or not to request extra assistance in implementing non-lethal control. If the rancher is in an area predicting high risk, they may enter into a cooperative agreement or if they are in an area with low risk continue to use their regular predator-prevention measures. In the Q&A session after the presentations, many of the panelists stressed the need for further scientific studies of the effects of lethal removal and the potential benefits of creating a risk map for Washington.
14
4. Rethink how lethal control is implemented.
The ecological research on social behavior in wolves (Brainerd et al.; 2008; Borg et al.; 2015; Smith, unpublished) provided information that could refine how lethal management is employed in Washington State. Currently, lethal removal of wolves in Washington State is done by USDA Wildlife Services through sharpshooters from helicopters, or by staff on the ground, who find the wolves by following a GPS collared wolf and killing wolves suspected to be problem animals. The age and sex of the wolves are identified after the wolves are shot. Given the current best science on wolf social structure, managers should consider means to allow for selective removal of identified culprits in cases of repeated depredations by the same pack in an area. As this may be extremely difficult, especially in cases where landscape features render individual wolves difficult to identify, the best option may be following Brainerd et al.’s (2008) time and location recommendations for decreasing the impact of lethal removal. It also may be beneficial to revise the requirements concerning non-lethal measure implementation, focusing on designing site- specific preventative plans and making it obligatory to have those plans in place before depredations occur in order to consider lethal removal, similar to the ODFW system (ODFW, 2013). The continued implementation and refinement of the WDFW’s conflict specialist program can aid in coordinating with ranchers on pre-depredation preventative measures. In addition, a redefined temporal scale on the wolf-management checklist showing the amount of time non-lethal measures should be in place between depredations before lethal control is employed could provide the time to consider ecologically-based management options without the immediate pressure to remove wolves.
With the creation of a risk map, it may also be possible to establish targeted areas with a high probability of livestock depredation where more decisive implementation of lethal removal could be beneficial. However, the work by Brainerd et al. (2008) suggests that, to the extent possible, lethal removal should be limited to solitary individuals and territorial pairs, wolf packs that are large with older pups, wolf packs that are close to other packs, and when it is not breeding season (Borg et al., 2015).
5. Understand the human dimension.
Dr. Martorello emphasized that one of the biggest struggles for the WDFW was a gap in understanding the human dimensions of wolf recovery. The ecological recovery of wolves may
15 be progressing, but increasing efforts will need to be made to understand and work within the social constraints of wolf management. Overall, there is a marked lack of trust between the management agencies, conservation groups, and livestock producers in Washington. Dr. Bruskotter’s research on the importance of elite cues in identity conflicts may be helpful in informing efforts to address this issue and foster coexistence. For example, the use of lethal control and lack of mandatory use of non-lethal control against wolves prior to depredations may send cues to the livestock producers that lethal control is efficient and endorsed. In contrast, efforts by state agencies to translocate problem animals (e.g. black bears) instead of shooting them could send cues to the public that the species being moved is important and wanted in the backcountry.
Most of the panelists highlighted the necessity of trust between the management agencies and the general public, stressing one-on-one efforts to build relationships, being transparent, and decreasing polarization by focusing on common goals. Both Dr. Bruskotter and Dr. Treves mentioned the importance of talking about the potential benefits of carnivores and reframing the ongoing conflict. Right now, much of the general public, as well as some natural resource managers, are asking the question, “How do we live with animals that can kill us and the things we value?” It may be that the way to move forward is to instead focus on the potential positives associated with wolves, and other carnivores, rather than on the conflict they can create. This is especially so given all of the trappings of modern life that people use and interface with daily which are far more likely to cause death (e.g., driving an automobile, falling from a ladder, etc.) yet which people readily accept and engage in without fear or forethought. The current frame surrounding carnivore management suggests a rational approach to dealing with animals, but it seems that our behavior is at least partially determined by our emotions. Additionally, a new frame and perspective may create spaces of common ground to promote conflict resolution between traditionally opposed groups (Asah et al., 2012). Because wildlife is a public resource, it is also important for management to focus on all the legal uses of wildlife and preserve natural resources for future generations.
16
Summary
Lethal removal can disrupt wolf pack dynamics, inhibiting recovery objectives in recolonizing populations, potentially increase livestock depredation, and negatively affect human attitudes towards wolves (Brainerd et al., 2008; Borg et al., 2015; Wielgus and Peebles, 2014; Treves et al., 2009; Treves and Bruskotter, 2014). Thus, understanding the effects of lethal control in Washington State will require rigorous ecological and social science research of the kind presented at this panel discussion. Based on the information provided, it seems that Washington State could benefit from the construction of a risk map for increased precision in targeting potential problem areas for wolf depredations (Treves et al., 2011). Furthermore, human dimensions research on wolf recovery in Washington State is needed in order to gain a more nuanced understanding of the ways in which individuals and groups are viewing wolves and the potential reasons behind those perceptions. Stakeholders must play an integral part in wolf management. Trust and relationship-building efforts between all the stakeholders is critical; transparency in management is necessary. The results from the panel show that the needs of recolonizing and isolated wolf packs must be carefully assessed when considering lethal removal. In order to implement the best available science, we must know more about the effectiveness of various preventative measures, the habits and trends of wolves within Washington, and the needs and desires of the people involved.
17
Cited Sources:
1. Asah, Stanley T., Bengston, David N., Wendt, Keith, Nelson, and Kristen C. (2012). Diagnostic Reframing of Intractable Environmental Problems: Case of a Contested Multiparty Public Land-use Conflict. Journal of Environmental Management. 108, 108- 119.
2. Brainerd, S. M., Andrén, H., Bangs, E. E., Bradley, E. H., Fontaine, J. A., Hall, W., Iliopoulos, Y., Jimenez, M. D., Jozwiak, E. A., Liberg, O., Mack, C. M., Meier, T. J., Niemeyer, C. C., Pedersen, H. C., Sand, H., Schultz, R. N., Smith, D. W., Wabakken, P., and Wydeven, A. P. (2008). The Effects of Breeder Loss on Wolves. Journal of Wildlife Management. 72, 89-98.
3. Borg, L. B., Brainerd, M. S., Meier T.J., and Prugh R. L. (2015). Impacts of Breeder Loss on Social Structure, Reproduction and Population Growth in a Social Canid. Journal of Animal Ecology. 84(1), 177-187.
4. Bruskotter T. J. (2013). The Pendulum Revisited: Social Conflict over Wolves and their Management in the Western United States. Wildlife Society Bulletin. 37(3), 674-679.
5. Bruskotter T. J., Schmidt R. H., and Teel T. L. (2007). Are Attitudes toward Wolves Changing? A Case Study in Utah. Biological Conservation. 139, 211-218. 6. Dickman, A. J. (2010). Complexities of conflict: the importance of considering social factors for effectively resolving human-wildlife conflict. Animal Conservation, 13, 5, 458-466. 7. Dietsch A. M. Teel, T. L, Manfredo, M. J., Jonker, S.A., and Pozzanghera S. (2011). State report for Washington from the research project entitled “Understanding People in Places.” Project report for the Washington Department of Fish and Wildlife. Fort Collins, CO: Colorado State University, Department of Human Dimensions of Natural Resources.
8. Hogberg, J., Treves, A., Shaw B., and Naughton-Treves, L. (2013). Attitudes Toward Wolves Since the First Regulated Wolf Hunting and Trapping Season in Wisconsin. Institute for Journalism and Natural Resources. Iron Mountain, MI.
9. Kellert, S. R. (1978). Attitudes and Characteristics of Hunters and Anti-hunters. Transactions of the North American Wildlife and Natural Resources Conference 43, 412- 423.
10. Lute, M.L., Bump, A., and Gore, M.L. (2014). Identity-Driven Differences in Stakeholder Concerns about Hunting Wolves. PLoS ONE 9(12), e114460.
11. MacNulty, D. R., Smith, D. W., Mech, D. L., Vucetich A. J., and Packer, C. (2012). Nonlinear Effects of Group Size on the Success of Wolves Hunting Elk. Behavioral Ecology. 23(1), 75-82.
18
12. MacNulty, D. R., Smith, D. W., Mech, D.L., and Eberly, L. E. (2009) Body Size and Predatory Performance in Wolves: Is Bigger Better? Journal of Animal Ecology. 78, 532- 539.
13. Mech, D.L. (1994). Buffer Zones of Territories of Gray Wolves as Regions of Intraspecific Strife. Journal of Mammology. (75)1, 199-202.
14. ------(1999). Alpha Status, Dominance, and Division of Labor in Wolf Packs. Canadian Journal of Zoology. (77)8, 1196-1203.
15. Naughton-Treves, L., Grossberg, R., and Treves, A. (2003). Paying for tolerance: The Impact of Livestock Depredation and Compensation Payments on Rural Citizens' Attitudes toward Wolves. Conservation Biology. 17, 1500-1511.
16. Ordiz, A., Bischof, R., and Swenson, J. E. (2013). Saving Large Carnivores, but Losing the Apex Predator? Biological Conservation. 168, 128-133.
17. Oregon Department of Fish and Wildlife (2005, updated 2010). Oregon Wolf Conservation and Management Plan. Salem, OR.
18. Oregon Fish and Wildlife (2013). Wolf Lethal Take Rules in Oregon. Livestock Producer Reference. Salem, OR.
19. Responsive Management (2014). Washington Residents’ Opinions on Bear and Wolf Management and Their Experiences with Wildlife that Cause Problems. A study conducted for the Washington Department of Fish and Wildlife. Olympia, WA.
20. Slagle, K. M., Bruskotter, J.T., and Wilson S. R. (2012). The Role of Affect in Public Support and Opposition to Wolf Management. Human Dimensions of Wildlife: an International Journal. 17(1), 44-57.
21. Treves A. and Bruskotter J. T. (2014). Tolerance for Predatory Wildlife. Science. 344, 476-477.
22. Treves A, and Karanth, K. U. (2003). Human-carnivore Conflict and Perspectives on Carnivore Management Worldwide. Conservation Biology. (17)6, 1491-1499.
23. Treves, A., Jurewicz, R., Naughton-Treves, L., and Wilcove, D. (2009). The Price of Tolerance: Wolf Damage Payments after Recovery. Biodiversity and Conservation. 18(14), 4003-4021.
24. Treves A., Martin A. K., Wydeven P.A., and Wiedenhoeft E. J. (2011). Forecasting Environmental Hazards and the Application of Risk Maps to Predator Attacks on Livestock. Bioscience. 61, 451-458.
25. Treves, A., Vuecetich, Cornman, and Mitchell (2013). Unpublished report to Michigan Natural Resources Committee available upon request.
19
26. Wallach, A. D., Ritchie, E. G., Read, J., and O’Neill, A. J. (2009). More than Mere Numbers: the Impact of Lethal Control on the Social Stability of a Top-Order Predator. PLoS One. 4(9), e6861.
27. Wielgus, R. B., and Peebles K. A. (2014). Effects of wolf mortality on livestock depredations. PLoS ONE. 9(12), e113505.
28. Wiles, G. J., Allen H. L., and Hayes G. E. (2011). Wolf Conservation and Management Plan for Washington Department of Fish and Wildlife. Olympia, WA. 29. (2013, January 1). Retrieved February 27, 2015, from http://www.thewildlifenews.com/ 30. (2015, February 24). Retrieved February 27, 2015, from http://www.dfw.state.or.us/wolves/
20
Appendix: Figure 1. WDFW decision process for implementing lethal control of wolves, Martorello, 2014.
21
Figure 2. Washington confirmed wolf pack locations and corresponding conflict specialists, Martorello, 2014.
22
Appendix E
COPIES OF NEW CITATIONS REFERENCED IN THIS JOINT CONSERVATION GROUPS’ DRAFT WOLF PLAN COMMENT LETTER TO CDFW BOREAL ENVIRONMENT RESEARCH 18 (suppl. A): 43-49 © 2013 ISSN 1239-6095 (print) ISSN 1797-2469 (online) Helsinki 18 December 2013
Wolves facilitate the recovery of browse-sensitive understory herbs in Wisconsin forests
Krystle Bouchard1l, Jane E. Wiedenhoeft2l, Adrian P. Wydeven2l and Thomas P. Rooney1l*
1> Department of Biological Sciences, 3640 Colonel Glenn Hwy., Dayton, OH 45435, USA (*corresponding author's e-mail: [email protected]) 2> Wisconsin Department of Natural Resources, Park Falls, WI 54452 USA
Received 1 Nov. 2012, final version received 25Apr. 2013, accepted 7 June 2013
Bouchard, K., Wiedenhoeft, J. E., Wydeven, A. P. & Rooney, T. P. 2013: Wolves facilitate the recovery of browse-sensitive understory herbs in Wisconsin forests. Boreal Env. Res. 18 (suppl. A): 43-49.
We asked whether wolf re-colonization would facilitate increased growth and reproduc tion of three browse-sensitive plant species. We hypothesized plant size and the proportion of reproductive individuals would be lowest in areas with no wolves, intermediate where wolves had been present for 4-6 years, and highest where wolves had been present for 12-13 years. Two plant species exhibited significantly greater reproduction where wolves were present for 12-13 years. Mean leaf size of indicator plants was significantly greater in areas where wolves were present for 12-13 years, as compared with that in areas where wolves were not present or were present for 4-6 years, but the effect size appears small. While the return of wolves to this region is likely to benefit browse-sensitive plant species, our findings suggest that wolf recovery will not generate a trophic cascade of sufficient magnitude to halt or reverse the loss of plant diversity in the Great Lakes region in the near term.
Introduction reduced growth and reproduction, truncated size structures, and population declines (Ander Following the extermination of predators and the son 1994, Rooney and Gross 2003, Balgooyen enactment of restrictive game laws, white-tailed and Waller 1995). Changes in plant community deer (Odocoileus virginianus) populations grew structure and composition following deer brows and their browsing resulted in major changes ing can indirectly alter composition of animal in forest community composition and structure assemblages, as is seen in birds (Allombert et al. throughout eastern North America (Cote et al. 2005, Martinet al. 2011). 2004). Deer browsing have been implicated in Wydeven et al. (2009) provide a brief his shifting community compositions reducing the tory of gray wolves (Canis lupus) in Wisconsin. abundance of palatable and non-resistant species Prior to European settlement, there was suf to less palatable and resistant species (Gi111992, ficient prey to support 3000-5000 wolves in Husheer et al. 2003, Rooney 2009). Repeated the state. Wolves were extirpated by 1960, and browsing of palatable and non-resistant her began recolonizing from Minnesota in the mid- baceous species can result in shorter stature, 1970s. Between 1980 and 2007, the wolf popu- Editor in charge of this article: Eero Nikinmaa 44 Bouchard eta/. • BOREAL ENV. RES. Vol. 18 (suppl. A) lation grew from 25-28 individuals to 540-577 Material and methods individuals (35.5 wolves per 1000 km2). Wolves are now widely distributed across the northern Site selection third of the state. Pack locations and sizes have been mapped and monitored using radio-collars Wolf pack locations have been mapped annu and winter track surveys since 1979 (Wydeven ally by the Wisconsin Department of Natural et al. 2009). The combination of extensive data Resources (WIDNR) since 1979, and are main and monitoring of a recovering wolf population, tained in a geographic information system. regionally-high deer densities, and impacted Annual shapefiles were overlayed in ArcGIS plant populations (Rooney et al. 2004) makes in order to determine how long an area was Wisconsin an ideal natural experiment for inves occupied by a wolf pack. Individual packs were tigating trophic cascades. selected for study based on a time criteria (either Studies from western North America have wolves continuously occupied the area for 4-6 demonstrated the effects recovering wolf popu years or 12-13 years). This created three treat lations can have on tree recruitment dynamics ments: no wolf impact, low wolf impact, and (Beschta and Ripple 2009). The reintroduction high wolf impact. Once a pack was selected, of wolves to Yellowstone National Park, for random points within its boundaries were chosen example, appears to have released aspen (Popu using ArcGJS and the coordinates recorded. lus tremuloides) and willow (Salix spp.) from Each potential site was visited and cruised to elk (Cervus elaphus) herbivory. The question determine that forest types were similar in age of whether wolves generate trophic cascades and composition, and contained populations of P. in midwestern forests is largely unexamined. pubescens. Polygonatum pubescens is common In years with high snowfall, wolves have influ throughout northern Wisconsin and was initially enced the growth rate of balsam fir (Abies bal used as a focal species. Where they co-occurred, samea) on Isle Royale by depressing moose we collected data from Clintonia borealis and abundance on the island (McLaren and Peterson, Trillium grandiflorum populations (measure 1994, Post et al. 1999). However, it is unclear ment details are provided below). However, the whether this trophic cascade is the exception or absence of one or both of these species did not the rule in mainland Wisconsin forests. constrain our site selection procedure. Once the In this study, we determined whether the re sites were deemed suitable, we randomly selected colonization of wolves could facilitate increased two packs from our list of wolf occupancy for growth and reproduction of browse-sensitive 4-6 years, and two packs from our list of wolf plant species. To do this, we compared vegeta occupancy 12-13 years. We established two sites tion in areas that had wolves for three different within the territorial boundary of each pack. periods of time. We compared areas without Wolf-free sites were selected in a manner wolves with areas that established wolf packs similar to sites with wolves. We used an ArcGIS for 4-6 years and for 12-13 years. We measured map overlay to identify and select potential non individual plant size, population size-structure, wolf sites. Potential non-wolf sites were selected and the proportion of reproductive individuals of in areas where the closest wolf pack boundary three herbaceous deer browse indicator species: was located at least 5 km away. Forest types were Polygonatum pubescens, Clintonia borealis, similar in age and composition, and contained and Trillium grandiflorum (Anderson 1994, Bal populations of P. pubescens. Four non-wolf sites gooyen and Waller 1995, Augustine and Frelich were chosen for inclusion in the experiment. 1998, Kirschbaum and Anacker 2005, Rooney and Anderson 2009). We hypothesized plant size, size structure variation, and the proportion Field methods of reproductive individuals would be lowest in areas with no wolves and highest where wolves Surveys were conducted in June 2008 and 2009. had been present for 12-13 years. A transect consisting of five 10 x 10 m plots BOREAL ENV. RES. Vol. 18 (suppl. A) • Wolves facilitate the recovery of understory herbs 45 each separated by 20m. Each transect was wolves" and "wolves present 4-6 years," "no located at least 30 m from the nearest road or wolves" and "wolves present 12-13 years," or ATV trail road, with most transects established "wolves present 4-6 years," and "wolves present parallel to a road or trail. Plots were systemati 12-13 years." In each case, we assume that areas cally sampled for P. pubescens until either all P. with wolves for a longer period of time reflect a pubescens plants were measured (see details greater wolf impact, and areas without wolves below), or alternatively, when 200 plants were or with wolves for a shorter period of time measured. We required a minimum of 50 indi reflect a lower wolf impact. We then computed viduals per site for analysis. Consequently, not the log response ratio L for the leaf area of each all sites were used in each analysis. A total of species where L = ln(Amore wolf impac/Aless wolr impact). 1268 P. pubescens were surveyed at eight sites When Amore wolf impact= Alesswolfimpact' L = 0. Negative (three no-wolf, two 4-6 year wolf, and three values of L indicate smaller plants in areas with 12-13 year wolf sites). We used the same pro less wolf impact, while positive values indicate cedure when sampling T. grandiflorum and C. larger plants in areas where wolves have been borealis. In total, we measured 476 T. grandi present for a longer period of time. A 95% confi florum total at three sites (1 per wolf occupancy dence interval (CI) was calculated for each spe treatment) and 558 C. borealis at 4 sites (two no cies L to determine if it differed from zero. wolf, one 4-6 year, and one 12-13 year). We combined results from all plant spe Within each plot, we counted the number of cies to examine the used techniques developed leaves for each P. pubescens and recorded if the for meta-analysis. Data from each species were plant was reproductive. The number of leaves combined to create a mean effect size, following per plant (x) is directly related to total leaf area, the procedures outlined in Hedges et al. (1999). y (y = 1.50x, r = 0.70, n = 49; Bouchard 2009), To account for among-species variation in effect so we used leaf count as a proxy for leaf area sizes, we combined effect sizes from each plant and hence plant size. We collected more detailed species to calculate the mean effect size, or measurements to estimate leaf area of T. gran overall effect. The effect size of each plant spe diflorum and C. borealis. For both species, the cies was first weighted by their inverse sampling length and width of each leaf (in mm), which variance plus a constant, q. The computation were converted into total leaf area using regres of q is derived from homogeneity analysis and sion analySiS (y =el.OOln(lengthxwidth)-0.58, r =0.99, n represents variability across population effects = 29 for T. grandiflorum; y = el.06ln(length X width)- 0.91' (Hedges et al. 1999). To determine if the mean r = 0.96, n =57 for C. borealis). The number of effect size differed from zero, we constructed reproductive plants was also tallied for T. gran 95% Cis. We considered top-down effects from diflorum and C. borealis. wolves statistically significant if 95% Cis did not include zero.
Data analysis Results To determine the top-down influence of wolves on plant reproduction, we pooled flowering data As compared with areas without wolves, plants across all sites within each wolf treatment. We growing in areas with wolves for a period of assessed differences among treatments were ana 4-6 years generally did not show any directional lyzed using Yates' X2 goodness of fit tests. trends. The mean size of P. pubsecens plants To determine the effects of wolf occupancy was 36% greater in the 4-6 year wolf treatments duration on average leaf area of each species, we than the non-wolf treatment. This difference was computed mean leaf area (A) in each wolf treat significant (p < 0.05; Fig. 1). However, the pro ment. Measurements for each plant species were portion of reproductive P. pubescens plants (27 pooled for each wolf treatment. We performed of 327, or 8.3%) in the 4-6 year wolf treatments analyses for three pairwise comparisons: "no was not significantly different than the propor- 46 Bouchard eta/. • BOREAL ENV. RES. Vol. 18 (suppl. A)
No wolves vs. wolves 4-6 yrs.
l\ '( <
('\, ·'' I I I' ., 11! ' '" . '
-1.0 -0.5 0.5 1.0
No wolves vs. wolves 12-13 yrs. Fig. 1 . Log response ratio (ratios of leaf area at sites with wolves for a longer time period relative to a shorter time period or wolf absence) and 95% confi dence intervals for all spe -1.0 -0.5 0.5 1.0 cies combined (thick line), Polygonatum pubescens Wolves 4-6 yrs. vs. wolves 12-13 yrs. (top thin line), C/intonia borealis (middle thin line}, • and Trillium grandiflorum (bottom thin line). Posi tive values indicate larger plants where wolves have been present for a longer period of time. Confidence intervals that intercept zero indicate no signifi -1.0 -0.5 0.0 0.5 1.0 cant difference (p > 0.05).
tion (37 of 479, or 7.2%) in the no-wolf areas (X2 the non-wolf treatment (Fig. 1). This result was =0.1, df = 1,p = 0.78). The mean leaf area of C. statistically significant. Additionally, the propor borealis plants was 3% smaller in the 4-<5 year tion of reproductive P. pubescens plants (79 of wolf treatments than the non-wolf treatment, but 433, or 18.2%) in the 12-13 year wolf treatments this difference was not statistically significant was more than twice the proportion (37 of 479, (Fig. 1). None of the 94 plants were reproduc or 7.2%) in the no-wolf areas (X2 = 21.7, df = 1, tive in the 4-<5 year wolf treatments, compared p < 0.001). Mean size of C. borealis plants was to zero of 301 in the non-wolf treatment. The 13% greater in the 12-13 year wolf treatments mean leaf area of T. grandiflorum plants was than the non-wolf treatment (Fig. 1). This result 30% smaller in the 4-<5 year wolf treatments than was statistically-significant. However, only 1 the non-wolf treatment, and this difference was plant of 200 was reproductive in the 12-13 year statistically significant (p < 0.05; Fig. 1). There wolf treatment. Zero of 338 plants was reproduc were no reproductive T. grandiflorum plants in tive in the no wolf treatment. Mean leaf area of the 4-<5 year wolf treatment, but 7.8% of the 191 T. grandiflorum plants in the 12-13 year wolf plants were reproductive in no-wolf areas (X2 = treatments was also 13% greater in than the non 12.7, df =1,p < 0.001). wolf treatment, but this was not significantly dif As compared with areas without wolves, ferent (Fig. 1). Of 191 T. grandiflorum plants in plants growing in areas with wolves for a period the 12-13 year wolf treatment, 7.3% (14 of 191) of 12-13 years showed some signs of recov were reproductive. This did not significantly ery. Mean size of P. pubsecens plants was 80% differ from the 7.8% reproductive (15 of 191) in greater in the 12-13 year wolf treatments than no-wolf areas (X 2 =0.0, df = 1 ,p = 1.0). BOREAL ENV. RES. Vol. 18 (suppl. A) • Wolves facilitate the recovery of understory herbs 47
As compared with areas with wolves for 4-6 were present for 12-13 years, as compared with years, plants growing in areas with wolves for that in areas where wolves were not present. The a period of 12-13 years were generally larger magnitude of the effect appears small. and more likely to flower. Mean size of P. pub Reproduction of browse-sensitive species secens plants was 30% greater in the 12-13 year usually declines in response to herbivory (Cote wolf treatments than the 4-6 year wolf treat et al. 2004, Kirschbaum and Anacker 2005), but ment (Fig. 1) and were 2.2 times more likely reproduction was a poor indicator of a response to flower (X2 = 14.66, df = 1, p < 0.001). Mean in this study. While over 1000 plants were sam size of C. borealis plants was 24% greater in the pled across the three species, we were unable 12-13 year wolf treatments than the 4-6 year to draw statistically-reliable conclusions about wolf treatment (p < 0.05; Fig. 1). Because only reproduction of C. borealis. Our other species a single plant was in flower, the influence of were most likely to flower in areas with wolves wolves on reproduction could not be assessed. for 12-13 years. Mean size of T. grandifiorum plants was 61% Meta-analysis of plant sizes indicated that greater in the 12-13 year wolf treatments than plants growing in the 12-13 year wolf treatment the 4-6 year wolf treatment (Fig. 1), and plants were significantly larger than plants growing in were more likely to flower (x2 = 5 .8, df = 1, p = the no-wolf treatment, but there was significant 0.02). NoT. grandifiorum plants flowered in the heterogeneity among species. There was an even 4-6 year wolf treatment. greater difference between growing the 12-13 Meta-analysis allowed us to combine the data year wolf treatment relative to plants growing in across species and examine the net effect. When the 4-6 year wolf treatment. Browse-sensitive the mean leaf sizes of plants were combined plant species performed most poorly at sites with into a single effect size, there was no significant wolves present for 4-6 years. Indeed, it appears effect of the 4-6 year wolf treatment on plant that plants growth and reproduction is higher size relative to areas without wolves (p > 0.05; with no wolves at all, relative to wolves present Fig 1). The combined effect size of all three for 4-6 years. Initially, this result puzzled us. In indicated a significantly larger leaf size when retrospect, however, we realized that our initial wolves were present for 12-13 years relative to hypothesis was faulty. When colonizing a new no wolves, as the lower bound of 95%CI was area, wolves select areas with high deer densities greater than zero (Fig. 1). The largest differences (Fuller 1989, Potvin et al. 2005). Our "no wolves" in mean leaf size were found between the 4-6 sites were probably located areas with fewer deer year wolf treatment and the 12-13 year wolf than elsewhere in the landscape. We would have treatment. The mean leaf size was significantly been wiser to sample vegetation in areas that greater in the 12-13 year wolf treatment for all had wolves for a brief period of time, such as three species, and the combined effect size was 1-2 years, instead of areas with no wolves. Such significant (Fig. 1). areas may have served more effectively as "high deer impact" sites within the wolf re-colonization chronosequence we wished to explore. Discussion The magnitude of plant recovery from deer browsing is much less than that found in the The re-colonization of the Great Lakes region aspen and cottonwood of Yellowstone after by wolves can be represented as a type of chron wolves were re-introduced (Ripple et al. 2001, osequence (Rooney and Anderson 2009). Packs Ripple and Beschta 2003, Beschta and Ripple became established in some areas 15 years ago, 2009). The differences in the magnitudes of while other areas have been colonized in the past vegetation response between our study and those few years. Time since re-colonization by wolves from Yellowstone could simply be the result of was associated with a modest increase in growth a few factors. Herbaceous plants growing in a and reproduction of browse-sensitive indicator shaded forest understory do not show the same plant species. Mean leaf size of indicator plants growth rate as woody species growing in sunny was significantly greater in areas where wolves riparian areas when released from herbivory. Elk 48 Bouchard eta/. • BOREAL ENV. RES. Vol. 18 (suppl. A) concentrate their foraging in discrete areas of bivores on forest ecosystems workshop." This research was the landscape with high quality forage. White supported by a Research Initiation Grant from Wright State University to TPR. tailed deer have high quality food distributed more evenly across the landscape, which could make their daily movements more unpredict References able to predators (Rooney and Anderson 2009). It is quite possible that trophic cascades are not Allombert S., Gaston AJ. & Martin J.L. 2005. A natural biologically important in the Great Lakes. Alter experiment on the impact of overabundant deer on song natively, 12-13 years is an insufficient amount bird populations. Biol. Conserv. 126: 1-13. of time for trophic cascades to become apparent. Anderson R.C. 1994. Height of white-flowered trillium (Trillium grandiftorum) as an index of deer browsing There were two key limitations of our study intensity. Ecol.Appl. 4: 104-109. design that could have affected our results. First, Augustine DJ. & Frelich L.E. 1998. Effects of white-tailed we did not statistically control for differences deer on populations of an understory forb in fragmented in wolf pack sizes in our study areas. Wolf deciduous forests. Conserv. Biol. 12: 995-1004. pack sizes are estimated every year, but pack Balgooyen C.P. & Waller D.M. 1995. The use of Clintonia borealis and other indicators to gauge impacts of white sizes change seasonally and from year to year. tailed deer on plant communities in northern Wisconsin, Between 1995-2007, mean pack size was 3.1 USA. Nat.Areas J. 15: 308-318. ± 0.3 wolves (Bouchard 2009). Second, we did Beschta R.L. & Ripple W.J. 2009. Large predators and not have good deer population density estimates trophic cascades in terrestrial ecosystems of the western from our study areas. Wisconsin estimates deer United States. Bioi. Conserv. 142: 2401-2414. Bouchard, KA. 2009. Finding the trophic trickle: using her densities for a deer management unit, and these baceous indicator species to investigate plant recovery units are -1000 km2 • Between 1995-2007, over from intense browsing by white-tailed deer (Odocoileus wintering densities averaged 11.2 ± 1.3 deer virginianus) after colonization by a top predator (Canis km-2 in the study area (Bouchard 2009). We have lupus). M.Sc. thesis, Wright State University. no information about deer density at the spatial Callan, R., Nibbelink, N.P., Rooney, T.P., Weidenhoeft, J.E., & Wydeven, A.P. 2013. Recolonizing wolves trigger a scale of study plots. Both limit the strength of trophic cascade in Wisconsin. J. Ecol. 101: 837-845. inferences we can draw. Cote S.D., Rooney T.P., Tremblay J.P., Dussault C. & Waller High deer densities throughout much of D.M. 2004. Ecological impacts of deer overabundance. the upper Great Lakes region continue pose a Ann. Rev. Ecol. Evol. Syst. 35: 113-147. Gill R.M.A. 1992.A review of damage by mammals in north challenge to conservation efforts. Deer brows temperate forests: 1. Deer. Forestry 65: 145-169. ing contributes to the erosion of plant diversity Hedges L.V., Gurevitch J. & Curtis P.S. 1999. The meta-anal (Rooney et al. 2004). This in turn could lead ysis of response ratios in experimental ecology. Ecology to additional indirect effects on insects, birds, 80: 1150-1156. and other species (Rooney and Waller 2003). Husheer S.W., Coomes D.A. & RobertsonA.W. 2003. Long term influences of introduced deer on the composition While the return of wolves to this region is likely and structure of New Zealand Nothofagus forests. For. to have modest benefits that accrue to plants Ecol.A1anage. 181:99-117 (Callan et al. 2013), our findings do not suggest Kirschbaum C.D. & Anacker B.L. 2005. The utility of that the current wolf population is sufficient Trillium and A1aianthemum as phyto-indicators of deer to halt or reverse the loss of plant diversity in impact in northwestern Pennsylvania, USA. For. Ecol. A1anage.217:54-66. the Great Lakes region in the near term. The Martin T.G.,Arcese P. & Scheerder N. 2011. Browsing down recovery of browse-sensitive understory herbs our natural heritage: deer impacts on vegetation structure in Wisconsin forests is dependent on the severity and songbird populations across an island archipelago. of previous browsing and the degree to which Biolog. Conserv. 144: 459-469. browsing is reduced (Kirschbaum and Anacker McLaren B.E. & Peterson R.O. 1994. Wolves, moose and tree rings on Isle Royale. Science 266: 1555-1558. 2005, Rooney et al. 2004). Post E., Peterson R.O., Stenseth, N.C. & McLaren, B.E. 1999. Ecosystem consequences of wolf behavioral Acknowledgements: We thank E. Nikinmaa and two anony response to climate. Nature 401: 905-907. mous referees for helpful comments on an earlier version Poszig D. & Theberge J.B. 2000. Gray wolf, Canis lupus this manuscript, and F. Vodde and K. Jogiste for organizing lycaon, responses to shifts of white tailed deer, Odocoi the 2012 "Impact of ungulates and other mammalian her- leus virginianus, adjacent to Algonquin Provincial Park, BOREAL ENV. RES. Vol. 18 (suppl. A) • Wolves facilitate the recovery of understory herbs 49
Ontario. Canadian Field Nat. 114: 62-71. 103-111. Potvin MJ., Drummer T.D., Vucetich JA., Beyer D.E., Rooney T.P. & Gross K. 2003 A demographic study of deer Peterson, R.O. & Hammill J.H. 2005. Monitoring and browsing impacts on Trillium grand(fiorum. Plant Ecol. habitat analysis for wolves in Upper Michigan. J. Wildt. 168: 267-277. A1anage.69: 1660-1669. Rooney T.P., Wiegmann S.M., Rogers DA. & Waller D.M. Ripple WJ., & Beschta R.L. 2003. Wolf reintroduction, 2004. Biotic impoverishment and homogenization in predation risk, and cottonwood recovery in Yellowstone unfragmented forest understory communities. Conserv. National Park. For. Ecol. A1anage.184: 299-313. Bioi. 18:787-798. Ripple W.J., Larsen E.J., Renkin RA. & Smith D.W. 2001. Wydeven A.P., Wiedenhoeft J.E., Schultz R.N., Thiel R.P., Trophic cascades among wolves, elk, and aspen on Yel Jurewicz R.L., Kohn B.E. & Van Deelen T.R. 2009. His lowstone National Park's northeru range. Bioi. Conserv. tory, population growth, and management of wolves in 102: 227-234. Wisconsin. In: Wydeven A.P., Van Deelen T.R. & Heske Rooney T.P. 2009. High white-tailed deer densities ben E.J. (eds.), Recovery of gray wolves in the Great Lakes efit graminoids and contribute to biotic homogeniza region of the United States: an endangered species suc tion of forest ground-layer vegetation. Plant Ecol. 202: cess story, Springer-Verlag, New York, pp. 87-106. CALIFORNIA FISH AND GAME COMMISSION NOTICE OF FINDINGS AND NOTICE OF PROPOSED RULEMAKING Gray Wolf (Canis lupus)
NOTICE IS HEREBY GIVEN that the California Fish and Game Commission (Commission), at its June 4, 2014 meeting in Fortuna, California, made a finding pursuant to California Fish and Game Code section 2075.5(e), that the petitioned action to add the gray wolf (Canis lupus) to the list of endangered species under the California Endangered Species Act, Fish & G. Code, § 2050 et seq. (CESA) is warranted.1 See also Cal. Code Regs., tit. 14, § 670.1, subd. (i)(1).
NOTICE IS ALSO GIVEN that the Commission, consistent with Fish and Game Code section 2075.5, proposes to amend Title 14, section 670.5, of the California Code of Regulations, to add the California gray wolf to the list of species designated as endangered under CESA. See also id., tit. 14, 670.1, subd. (j).
I. BACKGROUND AND PROCEDURAL HISTORY
On February 27, 2012, the Center for Biological Diversity (Center), Big Wildlife, the Environmental Protection Information Center, and Klamath-Siskiyou Wildlands Center petitioned (Petition) the Commission to list the gray wolf as an endangered species under CESA. Cal. Reg. Notice Register 2012, No. 15–Z, p. 494. The Commission received the Petition on March 12, 2012, and referred it to the Department of Fish and Wildlife (Department) for an initial evaluation on March 13, 2012. Cal. Reg. Notice Register 2012, No. 15-Z, p. 494. On June 20, 2012, the Commission granted a request by the Department for an additional thirty (30) days to complete its initial evaluation of the Petition.
On August 1, 2012, the Department submitted its Initial Evaluation of the Petition to List the Gray Wolf (Canis lupus) under the California Endangered Species Act (CESA) (August 1, 2012) (hereafter, the 2012 Candidacy Evaluation Report), recommending that the Petition provided sufficient information such that listing may be warranted and, therefore, that the Commission accept the Petition for further evaluation under CESA. Fish & G. Code, § 2073.5, subd. (a)(2); Cal. Code Regs., tit. 14, § 670.1, subd. (d).
On October 3, 2012, the Commission voted to accept the Petition for further evaluation and to initiate a review of the status of the species in California pursuant to Fish and Game Code section 2074.2, subdivision (e)(2). Upon publication of the Commission’s notice of determination, the gray wolf was designated a candidate species on November 2, 2012. Cal. Reg. Notice Register 2012, No. 44-Z, p. 1610 (2012 Candidacy Evaluation Report).
Consistent with the Fish and Game Code and controlling regulation, the Department commenced a 12-month status review of the gray wolf following published notice of its designation as a candidate species under CESA. As part of that effort, the Department solicited data, comments, and other information from interested members of the public and the scientific and academic community; and the Department submitted a preliminary draft of its status review for independent peer review by a number of independent reviewers who possess the knowledge and expertise to critique the validity of the report. Fish & G. Code, §§ 2074.4, 2074.8; Cal. Code Regs., tit. 14, § 670.1, subd. (f)(2).
1 The definition of an “endangered species” for purposes of CESA is found in Fish and Game Code section 2062. The effort culminated with the Department’s final Status Review of the gray wolf (Canis lupus) (February 5, 2014) (Status Review), which the Department submitted to the Commission at its meeting in Sacramento, California, on February 5, 2014. The Department recommended to the Commission that designating gray wolf as an endangered species under CESA is not warranted. Fish & G. Code, § 2074.6; Cal. Code Regs., tit. 14, § 670.1, subd. (f).
The Commission considered the Petition, the Department’s 2012 Candidacy Evaluation Report, the Department’s Status Review, and other information included in the Commission’s administrative record of proceedings at its meeting in Ventura, California on April 16, 2014, and at its meeting in Fortuna, California on June 4, 2014. Fish & G. Code, § 2075; Cal. Code Regs., tit. 14, § 670.1, subds. (g) and (i). After receiving additional information and oral testimony, the Commission determined, based on the requirements of CESA and the evidence before it, that listing gray wolf as an endangered species under CESA is warranted. Fish & G. Code, § 2075.5(a); Cal. Code Regs., tit. 14, § 670.1, subd. (i)(1)(A). In so doing, the Commission directed its staff to prepare findings of fact consistent with its determination for consideration and ratification by the Commission at a future meeting. The Commission also directed its staff, in coordination with the Department, to begin formal rulemaking under the California Government Code to add the gray wolf to the list of endangered species set forth in Title 14, section 670.5, of the California Code of Regulations. Fish & G. Code, § 2075.5(e)(2); Cal. Code Regs., tit. 14, § 670.1, subd. (j); Gov. Code, § 11340 et seq.
II. STATUTORY AND LEGAL FRAMEWORK
The Commission has prepared these findings as part of its final action under CESA to designate the gray wolf as an endangered species. As set forth above, the Commission’s determination that listing the gray wolf is warranted marks the end of formal administrative proceedings under CESA prescribed by the Fish and Game Code and controlling regulation. See generally Fish & G. Code, § 2070 et seq.; Cal. Code Regs., tit. 14, § 670.1. The Commission, as established by the California Constitution, has exclusive statutory authority under California law to designate endangered and threatened species under CESA. Cal. Const., art. IV, § 20, subd. (b); Fish & G. Code, § 2070.2
As set forth above, the CESA listing process for gray wolf began in the present case with the Center’s submittal of its Petition to the Commission in March 2012. Cal. Reg. Notice Register 2012, No. 15–Z, p. 494. The regulatory process that ensued is described above in some detail, along with related references to the Fish and Game Code and controlling regulation. The CESA listing process generally is also described in some detail in published appellate case law in California, including: Mountain Lion Foundation v. California Fish and Game Commission, 16 Cal.4th 105, 114–116 (1997); California Forestry Association v. California Fish and Game Commission, 156 Cal.App.4th 1535, 1541–1542 (2007);
2 Pursuant to this authority, standards, and procedures, the Commission may add, remove, uplist or downlist any plant or animal species to the list of endangered or threatened species, or to notice that any such species is a candidate for related action under CESA upon acceptance of a listing petition. Fish & G. Code, § 2074.2(a)(2); see also Cal. Code Regs., tit. 14, § 670.1, subd. (i)(1)(A)–(C). In practical terms, any of these actions may be commonly referred to as subject to CESA’s “listing” process.
2 Center for Biological Diversity v. California Fish and Game Commission, 166 Cal.App.4th 597, 600 (2008); and Natural Resources Defense Council v. California Fish and Game Commission, 28 Cal.App.4th 1104, 1111–1116 (1994).
The “is warranted” determination at issue here for the gray wolf is established by Fish and Game Code section 2075.5. Under this provision, the Commission is required to make one of two findings for a candidate species at the end of the CESA listing process; namely, whether the petitioned action is warranted or is not warranted. Here, with respect to gray wolf, the Commission made the finding under section 2075.5(2) that the petitioned action is warranted.
The Commission is guided in making this determination by the Fish and Game Code, CESA, other controlling law, and factual findings. The Fish and Game Code, for example, defines an endangered species under CESA as a “a native species or subspecies of a bird, mammal, fish, amphibian, reptile, or plant which is in serious danger of becoming extinct throughout all, or a significant portion, of its range due to one or more causes, including loss of habitat, change in habitat, overexploitation, predation, competition, or disease.” Fish & G. Code, § 2062. As established by published appellate case law in California, the term “range” for purposes of CESA means the range of the species within California. California Forestry Ass’n v. California Fish and Game Comm’n, supra, 156 Cal.App.4th at 1540, 1549-1551. The Fish and Game Code, CESA, and other controlling law do not require a species to have a continuous presence or a breeding population in California in order to meet the definition of “endangered” or “threatened.”
The Commission is also guided in making its determination regarding gray wolf by Title 14, section 670.1, subdivision (i)(1)(A), of the California Code of Regulations. This provision provides, in pertinent part, that a species shall be listed as endangered or threatened under CESA if the Commission determines that the species’ continued existence is in serious danger or is threatened by any one or any combination of the following factors: 1. Present or threatened modification or destruction of its habitat; 2. Overexploitation; 3. Predation; 4. Competition; 5. Disease; or 6. Other natural occurrences or human–related activities.
Likewise, the Commission is guided in its determination regarding the gray wolf by Fish and Game Code section 2070. This section provides that the Commission shall add or remove species from the list it establishes under CESA upon receipt of sufficient information that the action is warranted. As the Commission’s findings reflect, the gray wolf’s continued existence in California is in serious danger due to multiple threats.
Furthermore, CESA provides policy direction indicating that all state agencies, boards, and commissions shall seek to conserve endangered species and threatened species and shall utilize their authority in furtherance of the purposes of CESA. Fish & G. Code, § 2055. This policy direction does not compel a particular determination by the Commission in the CESA listing context. Yet, the Commission made its determination regarding gray wolf mindful of this
3 policy direction, acknowledging that “‘[l]aws providing for the conservation of natural resources’ such as the CESA ‘are of great remedial and public importance and thus should be construed liberally.’” California Forestry Ass’n v. California Fish and Game Comm’n, supra, 156 Cal.App.4th at 1545-1546 (citing San Bernardino Valley Audubon Society v. City of Moreno Valley, 44 Cal.App.4th 593, 601 (1996); Fish & G. Code, §§ 2051 and 2052).
Finally, in considering these factors, CESA and controlling regulation require the Commission to actively seek and consider related input from the public and any interested party. See, e.g., id. §§ 2071, 2074.4, 2078; Cal. Code Regs., tit. 14, § 670.1, subd. (h). The related notice requirements and public hearing opportunities before the Commission are also considerable. Fish & G. Code, §§ 2073.3, 2074, 2074.2, 2075, 2075.5 and 2078; Cal. Code Regs., tit. 14, § 670.1, subds. (c), (e), (g) and (i); see also Gov. Code, § 11120 et seq. All of these requirements are in addition to those proscribed for the Department in the CESA listing process, including an initial evaluation of the Petition and a related recommendation regarding candidacy, and a 12-month status review of the candidate species culminating with a report and recommendation to the Commission as to whether listing is warranted. Fish & G. Code, §§ 2073.4, 2073.5, 2074.4 and 2074.6; Cal. Code Regs., tit. 14, § 670.1, subds. (d), (f) and (h).
III. FACTUAL BASES FOR THE COMMISSION’S FINDING
CESA provides for the listing of either “native species or subspecies of a bird, mammal, fish, amphibian, reptile, or plant.” Fish and G. Code, §§ 2062 and 2067. The Petition, and the Commission’s finding, applies to the gray wolf in California.
The factual bases for the Commission’s finding that listing gray wolf as an endangered species under CESA is warranted are set forth in detail in the Commission’s administrative record of proceedings. Substantial evidence in the administrative record of proceedings in support of the Commission’s determination includes, but is not limited to, the Petition, the Department’s 2012 Candidacy Evaluation Report, the Department’s 2014 Status Review, and other information presented to the Commission and otherwise included in the Commission’s administrative record of proceedings as it existed up to and including the meeting in Fortuna, California on June 4, 2014. The Commission made its final determination under CESA with respect to gray wolf at that meeting. Fish & G. Code, § 2075; Cal. Code Regs., tit. 14, § 670.1, subds. (g) and (i).
The Commission finds the substantial evidence supports the Commission’s determination under CESA that the continued existence of gray wolf in the State of California is endangered by one or a combination of the following factors: 1. Overexploitation; 2. Predation; 3. Disease; 4. Other natural occurrences or human-related activities.
The Commission also finds that there is in the record of administrative proceedings substantial evidence to establish that designating the gray wolf as an endangered species under CESA is warranted. The following Commission findings highlight in more detail some but not all of the evidence in the administrative record of proceedings that support the Commission’s determination that the gray wolf is in serious danger of becoming extinct throughout all, or a significant portion, of its range:
4
1. It is likely that wolves historically occurred in California and were widely distributed in the State. Status Review at 10 (“While limited the available information suggests that wolves were distributed widely in California, particularly in the Klamath-Cascade Mountains, North Coast Range, Modoc Plateau, Sierra Nevada, Sacramento Valley, and San Francisco Bay Area. The genetic evidence from southeastern California suggests that the Mexican wolf may have occurred in California, at least as dispersing individuals. While the majority of historical records are not verifiable, for the purposes of this status review, the Department concludes that the gray wolf likely occurred in much of the areas depicted (CDFW 2011a) (Figure 1)); 2012 Candidacy Evaluation Report at 4 (“As to the science available at this time and the reasonable inferences that can be drawn from that information, it indicates to the Department at this time that wolves were likely broadly distributed in California historically … .”); id. at 10 (“In summary, historic anecdotal observations are most consistent with a hypothesis that wolves were not abundant, but widely distributed in California.”).
2. There is sufficient evidence to conclude that wolves occurred historically in California. However, by the late 1920’s, the species was extirpated from the state. Status Review at 4 (“2012 Candidacy Evaluation Report at 4) (“As to the science available at this time and the reasonable inferences that can be drawn from that information … humans likely purposefully extirpated the species in California early in the twentieth century.”)
3. Following listing of the gray wolf under the federal Endangered Species Act in 1974 and recovery efforts during the 1990s, a population of gray wolves in the Northern Rocky Mountain states has been re-established through a federal recovery program, and dispersing wolves from this population have established territories and several packs in Washington and Oregon. 2014 Status Review at 28.
4. In September 2011, a radio-collared, sub-adult gray wolf known as “OR7” dispersed from the Imnaha pack in northeastern Oregon and arrived in California on December 28, 2011, marking the first documented individual of the species in California since the 1920s. 2012 Candidacy Evaluation Report at 4 (“a single lone wolf, a dispersing young male named ‘OR7,’ entered California in December 2011, remaining largely in the State since that time”); id. at 10 (“The first gray wolf detected in California after many decades occurred in December 2011 with the arrival of ‘OR7,’ a radio-collared, sub- adult gray wolf that dispersed from a pack in Oregon.”); id. (“OR7 dispersed from the Northeastern Oregon’s Imnaha pack in September 2011.”)
5. The gray wolf is once again present in California, on at least an intermittent basis, and foreseeably will continue to be present in California, as discussed below. OR-7’s range now includes California and Oregon. OR7 has established a range that includes portions of Northern California, as this wolf is known to have crossed back and forth across the Oregon-California border since 2011 and to have been present in California in each of those years. Status Review at 4 (“The lone radio-collared gray wolf, OR7, dispersed from northeastern Oregon’s wolf population to California in December 2011 and has been near the Oregon/California border since that time, crossing back and forth.”); id. at 18 (“As far as the Department is aware, there is one gray wolf (OR7) that is near the Oregon/California border such that it may be in either state at any time.”); 2012 Candidacy Evaluation Report at 11 (“OR7 has passed back and forth over the
5 California/Oregon border several times over the last five months … .”); California Department of Fish and Wildlife, Gray Wolf OR7: Updates on wolves migrating to California (available at http://californiagraywolf.wordpress.com); see also Oregon Department of Fish and Wildlife, OR-7 Timeline of Events (available at http://www.dfw.state.or.us/wolves/OR-7.asp) (documenting OR7’s presence in California in each of 2011, 2012, 2013, and 2014).
6. OR7 has utilized areas of suitable habitat, primarily on public lands, comprised of ponderosa pine forests, mixed conifer forests, lava flows, sagebrush shrublands, juniper woodlands, as well as private lands including timberlands and agricultural lands, and has exhibited normal dispersal behavior for a young male gray wolf as he has sought to find other wolves, to establish his own pack, or to become part of an established wolf pack. 2012 Candidacy Evaluation Report at 10 (“It is believed that OR7 is exhibiting normal dispersal behavior for young male wolves, seeking to find other wolves, to establish his own pack, and/or to become part of an established wolf pack.”); id. at 11 (“OR7 has passed through ponderosa pine forests, mixed conifer forests, lava flows, sagebrush shrublands, juniper woodlands, and agricultural lands”); id. (“Although OR7 has used private lands (timberlands in particular), most of its route has traversed public lands.”).
7. On June 4, 2014, the State of Oregon Department of Fish and Wildlife confirmed that OR7 had mated with a female wolf of unknown origin, and that the pair was denning with a litter of at least two pups on public land in southwestern Oregon. See Press Release, Oregon Department of Fish and Wildlife, Pups for wolf OR7 (June 4, 2014) (“Wolf OR7 and a mate have produced offspring in southwest Oregon’s Cascade Mountains, wildlife biologists confirmed this week.”); Comments of Pamela Flick, Defenders of Wildlife (June 4, 2014 Commission hearing) (reporting breaking news that a remote camera in southwestern Oregon has detected at least two pups).
8. As the gestation period for gray wolves is 62-63 days and OR7 was documented in northern California on February 5, 2014, it is likely that OR7’s mate was traveling with OR7 in California at the time. Status Review at 10 (“The gestation period for wolves is 62-63 days.”); Testimony of Amaroq Weiss, June 4, 2014 Commission Meeting (Powerpoint slides at 15) (“A breeding population is likely on the border right now and a pregnant female was likely present in California already this year.”); L.D. Mech & L. Boitani, editors. Wolves: behavior, ecology, and conservation. University of Chicago Press, Chicago, Illinois, USA (cited in 2012 Candidacy Evaluation Report and Status Review) (discussing in Chapter 2 the reproductive behavior of wolves, and how wolves spend many months together leading up to impregnation and gestation).
9. The evidence in the record regarding wolf migration and dispersal behavior at a minimum indicates that wolves other than OR7 have similarly dispersed or will disperse to California, as most wolves from Oregon packs are not collared with radio transmitters and their presence in California may not otherwise have been detected (“we have acknowledged that we know of one [wolf, OR7]” and that “there could be others that we don’t know about”); U.S. Fish and Wildlife Service, Montana Fish, Wildlife & Parks, Nez Perce Tribe, National Park Service, Blackfeet Nation, Confederated Salish and Kootenai Tribes, Wind River Tribes, Washington Department
6 of Wildlife, Oregon Department of Wildlife, Utah Department of Wildlife Resources, and USDA Wildlife Services. 2011. Rocky Mountain Wolf Recovery 2010 Interagency Annual Report. C.A. Sime and E. E. Bangs, eds. USFWS, Ecological Services, 585 Shepard Way, Helena, Montana. 59601. (2011) at 2 (noting that “it is difficult to locate lone dispersing wolves.”); Carroll (2013) (Peer Review) at 5-6 (“[n]ot all Oregon wolves are detected and collared” so “it is possible that not all wolves dispersing to California have been detected”). Petition at 15 (“… it is impossible to rule out the possibility that previous dispersal events to California may … have occurred, which simply went un- detected because it is difficult to locate and track dispersing individual wolves”); Comments of Eric Loft (April 16, 2014 Commission Hearing).
10. The presence of wolves in California is small and is likely to remain small for the foreseeable future. Eisenberg (2013) (Peer Review) at 2 (“Any wolves becoming established in California will initially constitute a small population.”).
11. Dispersing wolves and small wolf populations are inherently at risk due to demographic and environmental stochasticity and in the case of wolves, of being killed by poachers, or hunters that mistake them for coyotes. Status Review at 5 (“A small population in California would be at some inherent risk although the species has demonstrated high potential to increase in other states. Dispersing individuals and small packs would likely be at highest risk due to population size.”); id. at 19 (“It is possible that a coyote hunter could mistake a gray wolf for a coyote, particularly at a long distance.”); id. at 22 (“With at least one gray wolf near the border of Oregon/California, and the knowledge that populations or species ranges are typically so large that they could range across both states …, an individual wolf, or a small number of wolves would be threatened in their ability to reproduce depending on the number and sex of the animals present in the range.”); 2012 Candidacy Evaluation Report at 6 (“Wolves are often confused with coyotes (Canis latrans) and domestic dogs (C. lupus familiaris), and wolf hybrids, which result from the mating of a wolf and a domestic dog.”).
12. Despite losses of areas of the gray wolf’s historic range in California, large tracts of habitat remain in the State that are sufficient to support a wolf population, particularly in the Modoc Plateau, Sierra Nevada, and Northern Coastal Mountains. Status Review at 17 (“Habitat Suitability Modeling: There are studies that have modeled potential suitable wolf habitat in California. Carroll (2001) modeled potential wolf occupancy in California using estimates of prey density, prey accessibility and security from human disturbance (road and human population density). Results suggested that areas located in the Modoc Plateau, Sierra Nevada, and the Northern Coastal Mountains could be potentially suitable habitat areas for wolves.
13. Since entering California, there have been threats to harm or kill OR7 or other wolves found in the State. (See e.g. May 6, 2013 Center for Biological Diversity letter to Department of Fish and Wildlife, p.13.) Although many people are supportive of gray wolves as a component of wildland ecosystems, wolves are considered a threat to livestock and wild ungulates by many other people, and are considered a threat to people by some. For example, the administrative record includes reports of statements by county supervisors from Modoc, Siskiyou, and Lassen counties expressing a desire to kill wolves in the area, a sentiment which represents an
7 imminent threat to wolves that are dispersing to the State. Status Review at 4-5 (“It is believed that limiting human-caused mortality through federal protection has been one of the key reasons that recovery efforts in the northern rocky mountains were successful.”); id. at 18-19 (“Public perception of wolf attacks on people, the documented losses of livestock, and the sometimes photographed killing of livestock or big game, continues to influence human attitudes toward wolves.”); Lassen County Board of Supervisors Hearing (Feb. 21, 2012) (quoting Lassen County supervisor to CDFW spokesperson) (“If I see an animal in my livestock, I kill it. If I kill a wolf, you going to throw me in jail? I don’t care what it is.”) (from notes taken at board meeting by Amaroq Weiss, Center for Biological Diversity); Modoc County Board of Supervisors Meeting (quoting Modoc County Supervisor) (“If I see a wolf, it’s dead.”) (Modoc County Board of Supervisors January 24, 2012 Hearing, Audio Archive); Chair of the Siskiyou County Board of Supervisors (“People are pretty much at their wits’ end trying to make a living with all the environmental protections that are being foisted upon them” and “we would like to see [wolves] shot on sight”) (Los Angeles Times (Dec. 24, 2011)) (available at http://articles.latimes.com/2011/dec/24/local/la-me-wolf- oregon-20111225).The Commission considers these statements and others like them to be compelling evidence of a threat to the continued existence of gray wolf in California. In a small early population of the species, loss of even one individual from human causes could significantly impact the ability of the species to thrive for years to come. CESA would criminalize such behavior in a more significant way than currently exists and act as a deterrent that may assist in allowing the early members of California’s gray wolf population to persist.
14. Humans are the primary factor in the past decline of wolves in the conterminous United States, including California, and humans remain the largest cause of wolf mortality as a whole in the western United States. Humans impact wolf populations through intentional predation (shooting or trapping) for sport or for protection; through unintentional killing, as gray wolves are often confused with coyotes (Canis latrans), domestic dogs (C. lupus familiaris), and wolf hybrids; through vehicle collisions; and through exposures to diseases from domestic animals.For example, the administrative record demonstrates that on more than one occasion, staff from the California Department of Fish and Wildlife have been fearful that OR7 and other unknown wolves that could be in California would be mistaken for a coyote and shot or harmed. Limiting human-caused mortality through federal protection has been one of the key reasons that the recovery effort in the Northern Rocky Mountains has been successful. Status Review at 4-5 (“It is believed that limiting human-caused mortality through federal protection has been one of the key reasons that recovery efforts in the northern rocky mountains were successful.”); id. at 19 (“Human-caused mortality of wolves is the primary factor that can significantly affect wolf populations (USFWS 2000, Mitchell et al. 2008, Murray et al. 2010, Smith et al. 2010)”); id. at 20.
15. Gray wolves are susceptible to several diseases including canine parvovirus and canine distemper, which has been responsible for extremely high rates of wolf pup mortality and suppression of wolf populations and which can be contracted from domestic dogs. Wolves are also susceptible to mange; mange-associated wolf population declines in Yellowstone National Park have led to pack extinction. Status Review at 23 (Wolves are vulnerable to a number of diseases and parasites, including, mange, mites, ticks, fleas, roundworm, tape worm, flatworm, distemper,
8 parvovirus, cataracts, arthritis, cancer, rickets, pneumonia, and Lyme disease.”); id. (“The transmission of disease from domestic dogs, e.g. parvovirus, is a grave conservation concern for recovering wolf populations (Paquet and Carbyn 2003; Smith and Almberg 2007). Recently, two wolves and two pups in Oregon were found to have died from parvovirus (ODFW 2013b). The disease is not thought to significantly impact large wolf populations, but it may hinder the recovery of small populations (Mech and Goyal 1993).”); id. (“Canine distemper and canine infectious hepatitis: Both diseases are known to occur in wolves and more recently canine parvovirus has become prevalent in several wolf populations (Brand et al. 1995)”); E.S. Almberg, P.C. Cross, A.P. Dobson, D.W. Smith and P.J. Hudson. 2012. Parasite invasion following host reintroduction: a case study of Yellowstone’s wolves. Philosophical Transactions of the Royal Society Bulletin. 367, p. 2840-2851).”).
16. Listing the gray wolf under CESA will allow the species to benefit from CESA’s protections, and would further the intent of the Legislature and be consistent with the objectives of CESA, i.e., the conservation, protection, restoration, and enhancement of species in their range in California. Protecting the gray wolf under CESA will also strengthen the Department’s existing stakeholder process to develop a state wolf plan, by providing clarity as to the management tools and options that will be available to the Department and to stakeholders. Status Review at 33 (“If the gray wolf species is listed under CESA, it may increase the likelihood that State and Federal land and resource management agencies will allocate funds towards protection and recovery actions.”); Carroll (2013) (Peer Review) at 6 (“Rather than using a dubious interpretation of CESA to decline to list a species due to its temporary and uncertain absence from state, California should follow the example of Washington and Oregon in using the relevant state statutes to protect colonizing wolves while at the same time developing multi-stakeholder plans that proactively restore wolf conservation and management issues.”).
IV. FINAL DETERMINATION BY THE COMMISSION
The Commission has weighed and evaluated the evidence presented for and against designating gray wolf as an endangered species under CESA. This information includes the Petition; the Department’s Petition Evaluation Report; the Department’s status review; the Department’s related recommendations; written and oral comments received from members of the public, the regulated community, various public agencies, and the scientific community; and other evidence included in the Commission’s record of proceedings. Based upon the evidence in the record the Commission has determined that the best information available indicates that the continued existence of the one or more gray wolves in California is in serious danger of extinction or threatened by present or threatened overexploitation, predation, disease, or other natural occurrences or human-related activities, where such factors are considered individually or in combination. (See generally Cal. Code Regs., tit. 14, § 670.1, subd. (i)(1)(A); Fish & G. Code, §§ 2062, 2067.) The Commission determines that there is sufficient evidence in the record to indicate that designating the gray wolf as an endangered species under CESA is warranted at this time and, with the adoption and publication of these findings and further proceedings under the California Administrative Procedure Act, the gray wolf shall be listed as endangered. See Cal. Code Regs., tit. 14, § 670.1, subd. (i)(1)(A).
9 Journal of Ecology 2013, 101, 837–845 doi: 10.1111/1365-2745.12095 Recolonizing wolves trigger a trophic cascade in Wisconsin (USA)
Ramana Callan1*, Nathan P. Nibbelink2, Thomas P. Rooney3, Jane E. Wiedenhoeft4 and Adrian P. Wydeven4 1 State University of New York College of Environmental Science and Forestry, Box 37, Wanakena, NY 13695, USA; 2 Warnell School of Forestry and Natural Resources, The University of Georgia, 180 E. Green Street, Athens, GA 30602, USA; 3Department of Biological Sciences, Wright State University, 3640 Colonel Glenn Hwy, Dayton, OH 45435, USA; and 4Wisconsin Department of Natural Resources, 875 S 4th Street, Park Falls, WI 54552, USA
Summary 1. We tested the hypothesis that wolves are reducing local browse intensity by white-tailed deer, thus indirectly mitigating the biotic impoverishment of understorey plant communities in northern Wisconsin. 2. To assess the potential for such a top-down trophic cascade response, we developed a spatially and temporally explicit model of wolf territory occupancy based on three decades of wolf monitor- ing data. Using a nested multiscale vegetation survey protocol, we compared the understorey plant communities of northern white cedar wetlands found in high wolf areas with control sites found in low wolf areas. 3. We fit species–area curves for plant species grouped by vegetation growth form (based on their predicted response to release from herbivory, i.e. tree, seedling, shrub, forb, grass, sedge or fern) and duration of wolf territory occupancy. 4. As predicted for a trophic cascade response, forb species richness at local scales (10 m2) was sig- nificantly higher in high wolf areas (high wolf areas: 10.7 0.9, N = 16, low wolf areas: 7.5 0.9, N = 16, P < 0.001), as was shrub species richness (high wolf areas: 4.4 0.4, N = 16, low wolf areas: 3.2 0.5, N = 16, P < 0.001). Also as predicted, percentage cover of ferns was lower in high wolf areas (high wolf areas: 6.2 2.1, N = 16, low wolf areas: 11.6 5.3, N = 16, P < 0.05). 5. Beta richness was similar between high and low wolf areas, supporting earlier assumptions that deer herbivory impacts plant species richness primarily at local scales. Sampling at multiple spatial scales revealed that changes in species richness were not consistent across scales nor among vegeta- tion growth forms: forbs showed a stronger response at finer scales (1–100 m2), while shrubs showed a response across relatively broader scales (10–1000 m2). 6. Synthesis. Our results are consistent with hypothesized trophic effects on understorey plant communities triggered by a keystone predator recovering from regional extinction. In addition, we identified the response variables and spatial scales appropriate for detecting such differences in plant species composition. This study represents the first published evidence of a trophic cascade triggered by wolf recovery in the Great Lakes region. Key-words: biodiversity, determinants of plant community diversity and structure, grey wolf (Canis lupus), herbivory, Northern White Cedar (Thuja occidentalis), species–area relationship, trophic cascades, vegetation dynamics, white-tailed deer (Odocoileus virginianus), Wisconsin
(Paine 1980; Carpenter, Kitchell & Hodgson 1985). Such Introduction interactions are frequently used to justify carnivore conserva- Indirect interactions between carnivores and plants, mediated tion, despite limited experimental evidence of trophic by herbivores, are commonly referred to as trophic cascades cascades involving large mammalian predators (Ray 2005; Ripple, Rooney & Beschta 2010). Recent attempts to infer *Correspondence author. E-mail: [email protected] top-down effects of predators have drawn on comparisons
© 2013 The Authors. Journal of Ecology © 2013 British Ecological Society 838 R. Callan et al. across areas with and without predators (Berger et al. 2001; triggered by the recovery of the Great Lakes wolf population Terborgh et al. 2006), or correlative studies of vegetation in northern Wisconsin. In addition, by assessing community- response following predator reintroduction (Ripple & Beschta level responses as opposed to species-level responses and by 2012). One of the most well-known examples of a terrestrial measuring across several spatial scales of observation, we trophic cascade is the wolf (Canis lupus)/moose (Alces alces)/ hope to inform future research by identifying the ideal balsam fir(Abies balsamea) system on Isle Royale (McLaren response variable and spatial scale for detecting effects of top & Peterson 1994). Despite its historical significance, cause predators in similar terrestrial systems. and effect in the Isle Royale system remains speculative due to the studies’ correlative nature and lack of replication or Ecological setting comparable control sites (Eberhardt 1997; Schmitz, Hamback & Beckerman 2000). Unlike in Yellowstone, where elk (Cervus elaphus) are the In addition, previously documented trophic cascades in primary prey species of grey wolves, wolves in the Great temperate terrestrial systems represent species-level as Lakes region prey mainly on white-tailed deer. Wolves were opposed to community-level cascades (Polis 1999). These nearly eradicated from the region during the early part of the studies tested how predators affect the productivity of one or 20th century. However, the wolf population in neighbouring occasionally several plant species (McLaren & Peterson 1994; Minnesota was never fully extirpated and began to recover Berger et al. 2001; Ripple & Beschta 2012), but failed to test under the protection of the Endangered Species Act. Dispers- whether predator manipulations affect species composition ing wolves from Minnesota (and possibly Canada) first began and diversity of entire plant communities. It has been argued to arrive in Wisconsin in the mid-1970s. The Wisconsin pop- that terrestrial cascades are principally species-level phenom- ulation grew slowly for the first few decades and then began ena, due to comparatively nonlinear food-web structure, to grow almost exponentially (Wydeven, Schultz & Thiel trophic complexity and effective plant defence mechanisms 1995; Wydeven et al. 2009). Wolf recovery in the Great (Halaj & Wise 2001). However, recent evidence from experi- Lakes region over the past three decades has been closely mental manipulations of herbivores and carnivores in old field monitored by the respective Departments of Natural ecosystems suggests that predators in terrestrial systems have Resources (DNR) in Minnesota, Wisconsin and Michigan. much stronger effects on plant species diversity than on plant The Wisconsin DNR (WiDNR) has annually mapped all biomass (Schmitz 2006). Furthermore, it is these changes known wolf pack territories in the state since 1979. The high in plant community composition that influence ecosystem quality of this data set provided the information we needed properties. to examine the spatial and temporal patterns in wolf occu- Ecological processes (including trophic cascades) are likely pancy throughout the state and thus answer the following to manifest differentially over a range of spatial and temporal question: Is the recovery of wolves releasing some understo- scales (Levin 1992; Polis 1999; Bowyer & Kie 2006). Size, rey plant communities from over-browsing by white-tailed generation time, reproductive characteristics and dispersal deer? ability of the organisms involved determine the scale(s) at Aldo Leopold reported irruptions (abrupt population rise) which they perceive and respond to environmental change of deer in Wisconsin as early as 1947 (Leopold, Sowls & (Levin & Pacala 1997). Variation in these life-history traits Spencer 1947). Based on land cover conditions, pre-settlement necessitates sampling at multiple spatial scales to accurately white-tailed deer densities in northern Wisconsin are thought interpret responses to top-down processes. Additionally, the to have ranged between 4 and 6 km 2 (McCaffery 1995). effects of trophic cascades are likely to be dampened by spa- The combination of predator extirpation, protective hunting tial heterogeneity (van Nes & Scheffer 2005). Habitat refugia laws and habitat management has contributed to current deer combined with spatial and temporal variability in species’ densities ranging between 4 and 15 km 2 (WiDNR 2010). distributions allow prey to escape predation (Halaj & Wise Densities as low as 1–2 deer km 2 have been prescribed 2001), potentially creating a mosaic of cascade intensity to improve recruitment of browse-sensitive plant species across the landscape. (Alverson, Waller & Solheim 1988). Is the recovering wolf The impacts of hyperabundant white-tailed deer (Odocoile- population in Wisconsin even capable of maintaining deer us virginianus) populations on understorey plant community densities this low? structure and composition are well-established (Alverson, In the Great Lakes region, wolves require 15–18 deer Waller & Solheim 1988; Tilghman 1989; Peek & Stahl 1997; ‘equivalents’ per wolf per year (Fuller 1989). Hence, the cur- Crete 1999; Rooney 2001; Rooney & Waller 2003; Horsley, rent Wisconsin wolf population, which has grown to c. 690 Stout & deCalesta 2003; Rooney et al. 2004; Holmes, Curran individuals (in winter) since their placement on the endan- & Hall 2008). However, few studies have examined how the gered species list (Wydeven & Wiedenhoeft 2010), has the recovery of wolves might moderate these effects. Recent capacity to take c. 12 000 deer per year. Given the current studies of species interactions in Yellowstone National Park estimated deer population of 340 000 in the Northern Forests (YNP) suggest that the recovery of wolf populations can natu- of Wisconsin (WiDNR 2010), region-wide effects of wolf rally ameliorate ungulate-caused ecosystem simplification recovery on deer populations are unlikely to manifest in the (White & Garrott 2005; Ripple & Beschta 2012). In this short term. In addition, whether wolf kills represent primarily study, we examine whether a similar trophic cascade was compensatory or additive mortality for white-tailed deer is in
© 2013 The Authors. Journal of Ecology © 2013 British Ecological Society, Journal of Ecology, 101, 837–845 Recolonizing wolves trigger a trophic cascade 839 part dependent on stochastic environmental variables such as winter severity (Mech & Peterson 2003). However, localized Wolves influences of wolf predation on deer populations are more probable, and drastic local herd reductions have been observed in Minnesota (Nelson & Mech 2006). + Hoskinson & Mech (1976) found higher white-tailed deer – survival on the edges of wolf territories as compared to their centres. Wolves are less likely to hunt in these buffer zones so as to avoid potentially fatal encounters with neighbouring + – wolf packs (Mech 1977). At local scales, the distribution of + + deer in northeastern Minnesota was found to be negatively correlated with wolf territory extents, and deer were found – White-tailed Deer primarily in buffer zones (Lewis & Murray 1993). Thus, buffer zones surrounding wolf pack territories may act as – refugia for white-tailed deer (Mech 1994). By extension, consistently occupied wolf pack territories may act as refugia Forbs Seedlings Shrubs Grasses Ferns Sedges for understorey plants that are preferred by white-tailed deer. Understory Plants Exclosure studies combined with resampling of historic vegetation plots from the 1950s (Curtis 1959) strongly impli- Fig. 1. Diagram of hypothesized tri-trophic interactions in northern cate the hyperabundance of white-tailed deer as the causal Wisconsin forests. Solid arrows represent direct positive and negative interactions. Dashed arrows represent hypothesized indirect interac- factor driving local losses in plant species diversity (Rooney tions. Dotted line represents competitive interactions. & Waller 2003; Rooney et al. 2004). This decline in rare and uncommon species is contributing to the biotic homoge- nization of understorey plant communities in northern Wisconsin (Frelich & Lorimer 1985; Cote et al. 2004; We anticipated that understorey plants would vary in their Rooney et al. 2004; Wiegmann & Waller 2006). Consistent response to release from browsing pressure dependent on the with this pattern, populations of northern white cedar (Thuja vegetation growth form in question. For example, tree seed- occidentalis) have suffered region-wide recruitment failure lings, shrubs and forbs are highly preferred by white-tailed due primarily to decades of over-browsing (Rooney, Solheim deer and collectively respond negatively to high browsing & Waller 2002). pressure. In contrast, ferns, grasses and sedges are generally White cedar forests are used intensively by white-tailed avoided by white-tailed deer and thought to respond posi- deer during the winter months, subjecting the highly nutri- tively (though indirectly) to high browsing pressure, because tious and palatable seedlings to excessive herbivory (Habeck they are released from competition with preferred species 1960; Van Deelen, Pregitzer & Haufler 1996). Historically, (Stromayer & Warren 1997; Cooke & Farrell 2001; Boucher, these coniferous wetlands have supported extremely diverse Cr^ete & Ouellet 2004). plant communities (Curtis 1959; Pregitzer 1990) providing Based on previous studies of deer influence on terrestrial habitat for a variety of rare lilies and orchids (USDA Forest plant communities (Frelich & Lorimer 1985; Stromayer & Service 2004). Unique shrub and forb species restricted to Warren 1997; Cooke & Farrell 2001; Rooney & Waller 2003; conditions found in white cedar wetlands are also susceptible Boucher, Cr^ete & Ouellet 2004; Cote et al. 2004; Wiegmann to over-browsing. Without recruitment to the canopy, exist- & Waller 2006), we predicted that high wolf areas would be ing mature stands of white cedar may become increasingly subject to reduced browse pressure and thus be characterized isolated as older stands senesce, accelerating the associated by an increased percentage cover of forbs, shrubs and seed- loss of understorey plant species restricted to this unique lings. We further expected that ferns, grasses and sedges habitat type (Alverson, Waller & Solheim 1988; Cornett would demonstrate the opposite response to wolf recovery et al. 2000) via the process of ‘relaxation’ described by (decreased percentage cover in high wolf areas). The relation- Diamond (1972). ship between disturbance and species diversity described by Given that northern white cedar wetlands are very sensitive Denslow (1985) predicts that species richness of seedling, to herbivory and are heavily used by white-tailed deer, we shrub and forb species should be higher at high wolf areas anticipated that recovery from over-browsing would be more (since browsing pressure should be lower and closer to his- easily detected in these ecosystems than in other forest cover toric levels). types. Thus, the tri-trophic cascade that we are testing for is As noted previously, the spatial scale at which species comprised of wolves, white-tailed deer and understorey plant respond to ecological processes is determined by the life-history communities of northern white cedar wetlands (Fig. 1). The traits of each species and thus likely to vary significantly. To objective of this study was to develop species–area curves to address this issue, we measured species richness across a test whether differences in plant species richness occur range of spatial scales (0.01, 1.0, 10, 100, 400 and 1000 m2). between high and low wolf areas (as defined by years of wolf In this manner, we sought to identify the appropriate scale of pack occupancy). measurement for detecting responses to release from
© 2013 The Authors. Journal of Ecology © 2013 British Ecological Society, Journal of Ecology, 101, 837–845 840 R. Callan et al. herbivory (as measured by changes in species richness) for RESEARCH DESIGN each vegetation growth form. Wolf packs establish and occupy territories that are patchily distrib- uted across the landscape (Mladenoff, Sickley & Wydeven 1999). Materials and methods The effect of wolves on deer abundance and foraging behaviour is likely to be limited to locations continuously occupied by wolf packs. Presumably, the impact of wolves increases with the size of the pack STUDY SITE and the number of years the territory has been consistently occupied. Data were collected throughout the Chequamegon-Nicolet National Since pack size and territory extent vary from year to year, this Forest, as well as from various state and county forests spanning creates a mosaic of potential impact intensity across the landscape. seven counties in north-central Wisconsin (Fig. 2). The forests of WiDNR population estimates of wolves were ascertained by live- northern Wisconsin are transitional between deciduous forests to the trapping and radiotracking, howl surveys and winter track surveys south and boreal forests to the north (Pastor & Mladenoff 1992; (Wydeven, Schultz & Thiel 1995). Territory extents were delineated Mladenoff et al. 1993). Northern white cedar wetlands occupy 5% of using minimum convex polygons based on radiolocations of collared the forested landscape (WiDNR 1998). This community type develops wolves and other wolf sign (Wiedenhoeft & Wydeven 2005). on poorly drained sites with a slight through-flow of groundwater, Using ArcGIS, we overlaid current wolf territories with historic ter- producing elevated pH and nutrient richness of the soil (Black & ritory extents to delineate areas that have been continuously occupied Judziewicz 2008). Mature stands of white cedar are densely shaded for c. 10 years (high wolf areas) and areas that have essentially with nearly closed canopies. The combination of these characteristics remained unoccupied since wolf recolonization of the region (low provides the unique light regimes and soil chemistry required by wolf areas). Only sites within the Chequamegon-Nicolet National species restricted to this community type (see below). Forest, state forest or county forest boundaries were selected. We Co-dominant trees in white cedar wetlands include balsam fir used the Combined Data Systems (CDS) data for the Chequamegon- (Abies balsamea), yellow birch (Betula alleghaniensis) and black ash Nicolet National Forest (USDA 2001) and various state and county (Fraxinus nigra). Tag alder (Alnus incana subsp. rugosa), hollies forest data sets to select stands characterized as northern white cedar (Ilex mucronata and I. verticillata), hazelnuts (Corylus spp.) and hon- wetlands. White cedar stands within consistently occupied wolf terri- eysuckles (Lonicera spp.) are common understorey shrubs. Cedar tories were then paired with the closest unoccupied white cedar stand wetlands are rich in sedges (e.g. Carex disperma, C. trisperma), ferns of similar stand area and stand age. In this way, plots were assigned (e.g. Dryopteris and Gymnocarpium spp.) and numerous wildflowers. to either high wolf areas (8–10 years of recent wolf occupancy) or Common wildflowers are goldthread (Coptis trifolia), starflower low wolf areas (0–3 years of recent wolf occupancy). To control for (Trientalis borealis), wild sarsaparilla (Aralia nudicaulis), naked spatial autocorrelation and limit the potential for confounding miterwort (Mitella nuda), blue-bead lily (Clintonia borealis), variables to produce false associations, we paired high wolf sites with bunchberry (Cornus canadensis), Canada mayflower (Maianthemum low wolf sites within a few kilometres (Fig. 3). canadense) and trailing ‘subshrubs’ such as creeping snowberry (Gaultheria hispidula), dwarf red raspberry (Rubus pubescens) and VEGETATION SURVEYS twinflower (Linnea borealis). Orchids include yellow lady’s slipper fl (Cypripedium parvi orum), heart-leaved twayblade (Listera cordata), We randomly placed 1 vegetation plot within each pre-selected white lesser rattlesnake plantain (Goodyera repens) and blunt-leaved bog cedar stand and surveyed a total of 32 cedar stands (16 in low and 16 orchid (Platanthera obtusata). in high wolf areas). Fourteen plots were completed in 2008, and 18
Fig. 2. Study areas in northern Wisconsin. Black triangles indicate vegetation plots located in high wolf areas. White triangles represent vegetation plots in low wolf areas.
© 2013 The Authors. Journal of Ecology © 2013 British Ecological Society, Journal of Ecology, 101, 837–845 Recolonizing wolves trigger a trophic cascade 841
using Akaike Information Criterion (AICc) (Barnett & Stohlgren 2003). The power model has an equation of the form: S ¼ cAz eqn 1
where S represents the number of species, A represents the area, and c and z are constants. For this type of analysis, the power function is often manipulated to log–log form: logðSÞ¼zlogðAÞþlogðcÞ eqn 2
Calculation of c and z values, where c = species richness at one unit of area (a-richness) and z = the rate at which species richness increases with area (b-richness), allows us to predict the direction and magnitude of differences in species richness. We grouped species– area curves for low and high wolf sites (n = 16) to compare a- and b-richness between these two treatments. Species–area curves were generated for all vegetation growth forms separately (note that grass Fig. 3. Intensity of wolf impact based on 10 years (1998–2008) of and sedge species richness data are from the first year of the study wolf pack territory data (WiDNR). Years of occupancy represent the only and are based on a reduced sample size, n = 7). T-tests and = – duration of wolf pack tenure. High wolf areas 8 10 years of occu- 95% confidence intervals were used to determine significant pancy, low wolf areas = 0–3 years of occupancy. differences in c and z values as well as to indicate at which scales differences are most easily detected. plots were completed in 2009. Vegetation surveys followed the Caro- lina Vegetation Survey (CVS) protocol developed by Peet, Wentworth Results & White (1998). Plots consisted of 10 modules (10 9 10 m) in a 2 9 5 array (1000 m2 in total). Four of the 10 modules were sampled PERCENTAGE COVER BY STRATA intensively, while the remaining plots were surveyed for additional fi species occurrences only. Two corners in each of the intensive mod- We identi ed a total of 199 vascular plant species: 23 trees, ules were sampled for the presence of vascular plant species (trees, 31 shrubs, 98 forbs, 12 ferns, 5 fern allies, 16 sedges, 7 shrubs, seedlings, ferns, forbs, grasses and sedges) using a series of grasses, 2 vines, 1 rush and 4 non-native species (see Callan nested quadrats (increasing incrementally in size from 0.01 to 10 m2). 2010 for a complete list). In general, sites with high wolf Percentage cover data were estimated visually for each 100-m2 occupancy had a diverse understorey community with com- module based on the following cover classes: 0–1%, 1–2%, 2–5%, plex vertical structure (Fig. 4a). In contrast, low wolf occu- – – – – – – 5 10%, 10 25%, 25 50%, 50 75%, 75 95%, 95 100%. pancy sites were characterized by a very limited herbaceous fi Identi cation of forbs conforms to Black & Judziewicz (2008). All layer and almost no woody-browse (Fig. 4b). Some low wolf other plant species names conform to Gleason & Cronquist (1991). sites were characterized by an understorey dominated by ferns Due to extensive time requirements, species identification of grasses but still lacking in forbs, shrubs and tree seedlings. and sedges was discontinued for the second field season. Percentage cover of forbs was higher in high wolf areas (high wolf areas: 15.0 4.4%, N = 16, low wolf areas: 8.8 2.5%, DATA ANALYSIS N = 16, P < 0.05) as were shrub and tree seedling cover com- bined (high wolf areas: 11.2 4.3%, N = 16, low wolf areas: Percentage cover of all plant species in each growth form (tree, shrub, 6.1 2.1%, N = 16, P < 0.05), while cover of ferns was lower seedling, forb, fern, grass, sedge) was assigned the geometric mean of = the cover class to which they were visually assigned. Geometric mean (high wolf areas: 6.2 2.1%, N 16, low wolf areas: = < values for each of the four intensive modules were then averaged to 11.6 5.3%, N 16, P 0.05) (Fig. 5). Percentage cover of provide one value for each plot. Student’s t-tests were used to com- grasses was equivalent in low and high wolf areas (high wolf pare percentage plant cover between high and low wolf areas across areas: 0.50 0.22%, N = 16, low wolf areas: 0.59 0.50%, all vegetation growth forms. N = 16, P = 0.32), and sedge cover did not differ significantly Species richness at each scale (0.01, 1.0, 10, 100, 400 and (high wolf areas: 7.4 4.0, N = 16, low wolf areas: 4.5 1.8, 1000 m2) was calculated for each plot by averaging subsamples. The N = 16, P = 0.10). Percentage tree cover was very similar number of subsamples varied depending on the scale sampled (0.01– between high and low wolf areas (high wolf areas: 69.9 5.7%, 2 = 2 = – 2 = 10 m , n 8, 100 m , n 4, and 400 1000 m , n 1). Again, N = 16, low wolf areas: 71.2 7.4%, N = 16, P = 0.39). Student’s t-tests were used to compare species richness between high and low wolf areas and across all vegetation growth forms and spatial scales. The multiscale nested structure of the CVS protocol also SPECIES– AREA RELATIONSHIPS facilitates the construction of species–area curves. Species–area curves – describe the rate at which species richness increases as the total area When all species were included in the analysis, species area sampled increases (Rosenzweig 1995). We fit averaged species curves in high wolf areas tended towards higher alpha richness values to the power function to determine y-intercept and richness (c) for all species combined (Table 1), but this differ- slope values (c and z values). We chose the power model because it ence was not significant (P = 0.10). Beta richness (z) ranged was shown to outperform the exponential model when evaluated from 0.27–0.35 across all sites, but was similar between low
© 2013 The Authors. Journal of Ecology © 2013 British Ecological Society, Journal of Ecology, 101, 837–845 842 R. Callan et al.
fi < (a) ence was not signi cant (P 0.10). Again, beta richness was equivalent between high and low wolf areas across all vegeta- tion growth forms. As predicted for a trophic response, forb species richness at local scales (10 m2) was significantly higher in high wolf areas (high wolf areas: 10.7 0.9 N, low wolf areas: 7.5 0.9, N = 16, P < 0.0001), as was shrub species rich- ness [high wolf areas: 4.4 0.4, low wolf areas: 3.2 0.5, N = 16, P < 0.001(Fig. 6)]. Contrary to our expectations, species richness of ferns was higher at the 10-m2 scale (high wolf areas: 2.99 0.3, low wolf areas: 2.08 0.47, N = 16, P < 0.01). Species richness of sedges was higher in high wolf (b) areas at the smallest spatial scale measured, 0.01 m2 (high wolf areas: 0.47 0.16, low wolf areas: 0.23 0.14 N = 7, P < 0.05), but this pattern was based on a limited sample size and was not observed at other spatial scales. Species richness of trees, seedlings and grasses was similar between low and high wolf areas across all scales.
Discussion As predicted, percentage cover of forbs was 70% higher on average in high wolf areas, and species richness of forbs was Fig. 4. Photo pair of understorey vegetation within the Chequamegon- 43% higher (at the 10-m2 scale). Shrubs showed a similar pat- Nicolet National Forest, WI. (a) Shows a high wolf area (within the tern with 84% higher percentage cover for seedlings and Bootjack Lake pack territory) and (b) shows the paired low wolf area (in the buffer zone between the Bootjack Lake pack and the Miles shrubs grouped and 39% higher species richness for shrubs Lake pack). alone. Percentage cover of ferns was 47% lower in high wolf areas. Although we expected greater species richness of tree seedlings in high wolf/low deer impact sites (Tilghman 1989), 90 this pattern was not observed. The presence of seedling spe- 80 cies may be more related to proximity to seed sources (adults
70 in the canopy) and less related to browsing pressure. The
60 nearly equivalent percentage tree cover between high and low wolf areas eliminates the possibility that differences in light 50 availability are responsible for the observed differences in 40 percentage cover of the lower strata. Percent cover 30 The similarity in percentage cover of grasses in high and * 20 * * low wolf areas was inconsistent with our predictions for a 10 top-down trophic response since previous studies indicated an
0 indirect positive relationship between deer browsing pressure Tree Shrub Forb Grass Sedge Fern and the percentage cover of grass species. Almost all visual – Fig. 5. Average percentage cover of high and low wolf area plots estimates of grass cover fell in the same cover class: 0 1%. 2 2 across 6 vegetation growth forms (forbs, shrubs, trees, ferns, grasses This area represents c. 1m of a 100-m module. Percentage and sedges) with 95% confidence intervals. Values are averaged geo- cover of grasses and sedges may need to be estimated at finer metric means of cover classes (0–1%, 1–2%, 2–5%, 5–10%, 10–25%, scales than at the 100-m2 module. Evidence does suggest that – – – – 25 50%, 50 75%, 75 95%, 95 100%). Hatched bars represent high sedges may actually be more abundant in high wolf areas. It wolf areas. Asterisks indicate significant differences (P < 0.05). is possible that sedge species in northern white cedar swamps respond negatively to white-tailed deer grazing even though and high wolf areas. When species richness of understorey Carex spp. collectively have been shown to respond posi- plants was broken down into vegetation growth forms based tively in other vegetation types (Wiegmann & Waller 2006). on their hypothesized response to herbivory, differences The similarity in z values (beta richness) between high and between high and low wolf areas were more pronounced low wolf areas suggests that herbivory may have little or no (Table 1). Alpha richness of forbs was much higher in high impact on species turnover, habitat heterogeneity or mass wolf areas (P < 0.001) as was alpha richness of shrubs effects. Although we observed consistent differences at (P < 0.05). Surprisingly, alpha richness of ferns was in fact broader scales, these may be due to local differences propa- higher in high wolf areas (P < 0.05), and alpha richness of gating up through higher scales of observation. Reduced sedges tended to be higher in high wolf areas, but this differ- browse intensity limits the ability of a few browse-resistant
© 2013 The Authors. Journal of Ecology © 2013 British Ecological Society, Journal of Ecology, 101, 837–845 Recolonizing wolves trigger a trophic cascade 843
Table 1. Slope (z or beta richness), intercept (c or alpha richness) and correlation coefficient (r2) values by vegetation growth form for species– area curves of northern white cedar stands with low and high wolf occupancy. Values in parentheses represent 95% confidence intervals
Low wolf areas High wolf areas
zc r2 zc r2
All Species 0.32 (0.30–0.35) 8.98 (7.34–10.99) 0.94 0.32 (0.30–0.35) 10.82 (9.16–12.79) 0.96 Forbs 0.24 (0.22–0.26) 3.82 (3.33–4.39) 0.91 0.24 (0.22–0.26) 5.42 (4.84–6.05) 0.93 Shrubs 0.30 (0.29–0.31) 1.35 (1.20–1.51) 0.95 0.32 (0.31–0.33) 1.57 (1.40–1.76) 0.95 Seedlings 0.26 (0.25–0.27) 1.20 (1.04–1.37) 0.96 0.27 (0.25–0.29) 1.25 (1.00–1.54) 0.93 Trees 0.22 (0.21–0.23) 0.76 (0.60–0.93) 0.87 0.21 (0.19–0.23) 0.78 (0.63–0.95) 0.86 Ferns 0.22 (0.20–0.24) 0.94 (0.71–1.20) 0.81 0.24 (0.23–0.25) 1.24 (1.09–1.41) 0.94 Grasses 0.20 (0.16–0.24) 0.52 (0.42–0.62) 0.83 0.21 (0.18–0.24) 0.60 (0.36–0.92) 0.77 Sedges 0.23 (0.20–0.26) 1.37 (1.00–1.83) 0.86 0.21 (0.23–0.25) 1.80 (1.41–2.27) 0.91
Plant species richness is determined by linked processes that act differentially across small, intermediate and large spatial scales (Schmida & Wilson 1985). Species richness at small scales (< 1m2) is a consequence of direct competition and niche relations (variability in resource utilization and alloca- tion). At intermediate scales (1–100 m2), species richness is more a consequence of microhabitat heterogeneity promoting the coexistence of species with different habitat requirements. At scales beyond 100 m2, species richness is more likely deter- mined by immigration of seeds from source habitats (‘mass effect’ dynamics, Schmida & Whitaker 1981). At this scale, the extent to which the plant community is linked to the regional species pool becomes the dominant process determining local recruitment and ultimately species richness (Rogers et al. 2009). Had we surveyed at scales < 1000 m2, we might expect a point at which species richness between high and low wolf areas would converge. However, patch occupancy of cedar stands and metapopulation dynamics could become dominant processes at this scale, superseding species–area relationships and strengthening or weakening differences in species richness values between high and low wolf areas. By sampling at multiple scales, we revealed that our ability to detect differences in species richness was not consistent among vegetation growth forms. Based on means and 95% confidence intervals, forbs show a stronger response at finer scales (1–100 m2), while shrubs show a response across broader scales (10–1000 m2). The design of future research should incorporate the proper scale in order to effectively – Fig. 6. Species area curves for high and low wolf areas displayed detect top-down effects. Many vegetation studies survey at across seven spatial scales for canopy trees, forbs and shrubs. High the scale of 1 m2, which is likely to miss significant differ- wolf area data points are represented by open circles and dashed lines. Data points are the mean number of species at each scale from ences in shrub species richness. Whether these scales are each plot. Scales on the y-axis (number of plant species) vary depend- appropriate for community types other than northern white ing on maximum species richness for each vegetation growth form. cedar wetlands is unknown. However, it is likely that the rele- X-axis intervals are not to scale. vant scales are determined by the process of deer herbivory itself and should be similar regardless of forest cover type. species to become locally dominant, thus increasing species Unfortunately, reciprocal relationships between trophic richness at local scales. Additionally, increased species rich- levels, like those found by McLaren & Peterson (1994) ness may be closely linked to increased density of individuals between wolves, moose and balsam fir on Isle Royale, are at local scales. This pattern has been observed in both lacking in Wisconsin. At present, deer data are available for temperate and tropical plant communities (Denslow 1995; the past several decades, but only at the very coarse scale of Busing & White 1997; Hubbell 2001; Schnitzer & Carson deer management blocks (WiDNR 2010). Since most low and 2001). high wolf areas in our study were within the same deer man-
© 2013 The Authors. Journal of Ecology © 2013 British Ecological Society, Journal of Ecology, 101, 837–845 844 R. Callan et al. agement unit, we considered existing deer data to be unsuit- manuscript benefitted from comments by Teresa Valentine, Chris Peterson, able for the scale of this study. Future research should focus Robert Cooper and Robert Warren. Fig. 1 was drawn by Mona Callan. on monitoring deer abundance and/or foraging behaviour concurrent with wolf occupancy and vegetation response. References Several factors that benefit both plant diversity and wolf hab- itat quality, irrespective of deer density and any sort of trophic Alverson, W.S., Waller, D.M. & Solheim, S.L. (1988) Forests too deer: edge effects in northern Wisconsin. Conservation Biology, 2, 348–358. effects, could result in the pattern that we documented. In par- Barnett, D. & Stohlgren, T.J. (2003) A nested intensity sampling design for ticular, road density has been shown to be negatively correlated plant diversity. Biodiversity and Conservation, 2, 255–278. with both plant diversity (Findlay & Houlahan 1997; Watkins Berger, J., Stacey, P.B., Bellis, L. & Johnson, M.P. (2001) A mammalian pred- ator-prey disequilibrium: how the extinction of grizzly bears and wolves et al. 2003) and wolf habitat selection (Mladenoff et al. 1995). affects the diversity of avian neotropical migrants. Ecological Applications, In addition, understorey vegetation in white cedar stands may 11, 947–960. fl be more influenced by hydrology and edge effects than by tro- Black, M.R. & Judziewicz, E.J. (2008) Wild owers of Wisconsin and the Great Lakes Region. The University of Wisconsin Press, Madison, Wisconsin, USA. phic effects. Landscape-level connectivity between cedar stands Boucher, S., Cr^ete, M. & Ouellet, J. (2004) Large-scale trophic interactions: is likely to influence mass effects. A bottom-up effect could white-tailed deer growth and forest understorey. Ecoscience, 11, 286–295. also be responsible for observed patterns. Areas with high plant Bowyer, R.T. & Kie, J.G. (2006) Effects of scale on interpreting life-history characteristics of ungulates and carnivores. Diversity and Distributions, 12, diversity may attract and maintain higher deer densities, which 244–257. in turn support successful establishment by wolf packs. Contin- Busing, R.T. & White, P.S. (1997) Species diversity and small-scale distur- ued research directed at ruling out confounding factors and dif- bance in an old-growth temperate forest: a consideration of gap partitioning concepts. Oikos, 78, 562–568. ferentiating between top-down and bottom-up effects is needed. Callan, R. (2010) Are Wolves Affecting the Biodiversity of Understorey Plant Our results provide compelling correlative evidence of Communities in Northern Wisconsin Via a Trophic Cascade? Doctoral Dis- top-down trophic effects generated by the recovery of sertation. University of Georgia, Athens, Georgia, USA. Carpenter, S.R., Kitchell, J.F. & Hodgson, J.R. (1985) Cascading trophic inter- Wisconsin’s wolf population. By addressing wolf impact at actions and lake productivity. BioScience, 35, 634–639. the scale of wolf territory extents, instead of presence/absence Cooke, A.S. & Farrell, L. (2001) Impact of muntjac deer (Muntiacus reevesi)at of wolves for entire regions, we were able to have both repli- Monks Wood National Nature Reserve, Cambridgeshire, eastern England. Forestry, 74, 241–250. cation of ‘treatments’ (n = 16) and comparable control sites Cornett, M.W., Frelich, L.E., Puettmann, K.J. & Reich, P.B. (2000) Conserva- (n = 16). We also identified species richness of forbs and tion implications of browsing by Odocoileus virginianus in remnant upland – shrubs in northern white cedar wetlands as ideal community- Thuja occidentalis forests. Biological Conservation, 93, 359 369. Cote, S.D., Rooney, T.P., Tremblay, J.P., Dussault, C. & Waller, D.M. (2004) level responses for detecting trophic cascades involving Ecological impacts of deer overabundance. Annual Review of Ecology, Evo- wolves and white-tailed deer in the boreal forests of the Great lution and Systematics, 35, 113–147. ^ Lakes region. Crete, M. (1999) The distribution of deer biomass in North America supports the hypothesis of exploitation ecosystems. Ecology Letters, 2, 223–227. The spatially hierarchical sampling design we developed to Curtis, J.T. (1959) The Vegetation of Wisconsin. University of Wisconsin Press, analyse wildlife census data in conjunction with vegetation Madison, Wisconsin, USA. data provides a template for addressing other broad-scale eco- Denslow, J.S. (1985) Disturbance-mediated coexistence of species. The Ecology of Natural Disturbance and Patch Dynamics (eds T.A. Pickett & P.S. logical impacts. Regardless of the process in question, multi- White), pp. 307–323. Academic Press, New York. scale approaches allow us to determine the scale at which a Denslow, J.S. (1995) Disturbance and diversity in tropical rain forests: the den- – pattern becomes detectable. The ability to detect such signals sity effect. Ecological applications, 5, 962 968. Diamond, J.M. (1972) Biogeographic kinetics: estimation of relaxation times above the ambient noise of ecological variation is essential to for avifaunas of southwest Pacific islands. Proceedings of the National Acad- understanding the relationship between pattern and process. emy of Sciences (USA), 69, 3199–3203. If the methods employed here were applied across other Eberhardt, L.L. (1997) Is wolf predation ratio-dependent? Canadian Journal of Zoology, 75, 1940–1944. forest types, we could predict long-term, region-wide effects Findlay, C.S. & Houlahan, J. (1997) Anthropogenic correlates of species of reintroducing top predators to this and other terrestrial sys- richness in Southeastern Ontario Wetlands. Conservation Biology, 11, 1000– tems. Our results indicate that wolf recovery in other regions 1009. Frelich, L.E. & Lorimer, C.G. (1985) Current and predicted long-term effects of North America (such as the northeastern United States) of deer browsing in hemlock forests in Michigan, USA. Biological Conserva- could be vital to maintaining the ecological integrity of north- tion, 34,99–120. ern white cedar wetlands (and potentially other temperate and Fuller, T.K. (1989) Population dynamics of wolves in North-Central Minnesota. Wildlife Monographs, 105,1–41. boreal forest systems as well). Whether efforts should be Gleason, H.A. & Cronquist, A. (1991) Manual of Vascular Plants of Northeast- focused on reintroducing wolves or on increasing the connec- ern U.S. and Adjacent Canada, 2nd edn. The New York Botanical Garden, tivity between existing wolf populations and unoccupied wolf Bronx, New York, USA. Habeck, J.R. (1960) Winter deer activity in the white cedar swamps of northern habitat should be carefully considered. Wisconsin. Ecology, 41, 327–333. Halaj, J. & Wise, D.H. (2001) Terrestrial trophic cascades: how much do they trickle? The American Naturalist, 157, 262–281. Acknowledgements Holmes, S.A., Curran, L.M. & Hall, K.R. (2008) White-tailed Deer (Odocoileus virginianus) alter herbaceous species richness in the Hiawatha National For- This research was made possible through a Graduate Research Assistantship est, Michigan, USA. American Midland Naturalist, 159,83–97. Award to R. Callan provided by the University of Georgia Graduate School Horsley, S.B., Stout, S.L. & deCalesta, D.S. (2003) White-tailed deer impact and the Warnell School of Forestry & Natural Resources. The University of on the vegetation dynamics of a northern hardwood forest. Ecological Appli- Wisconsin – Trout Lake Research Station provided an excellent home base for cations, 13,98–118. this work, and we especially thank S. Knight and Gretchen Hansen. We would Hoskinson, R.L. & Mech, L.D. (1976) White-tailed deer migration and its role like to thank Corey Raimond and Clare Frederick for their field assistance. The in wolf predation. Journal of Wildlife Management, 40, 429–441.
© 2013 The Authors. Journal of Ecology © 2013 British Ecological Society, Journal of Ecology, 101, 837–845 Recolonizing wolves trigger a trophic cascade 845
Hubbell, S.P. (2001) The Unified Neutral Theory of Biodiversity and Biogeog- Rooney, T.P. (2001) Deer impacts on forest ecosystems: a North American per- raphy. Princeton University Press, New Jersey. spective. Forestry, 74, 201–208. Leopold, A., Sowls, L.K. & Spencer, D.L. (1947) A survey of over-populated Rooney, T.P., Solheim, S.L. & Waller, D.M. (2002) Factors influencing the deer ranges in the United States. Journal of Wildlife Management, 11, 162– regeneration of northern white cedar in lowland forests of the Upper Great 183. Lakes region, USA. Forest Ecology and Management, 163, 119–130. Levin, S.A. (1992) The problem of pattern and scale in ecology. Ecology, 73, Rooney, T.P. & Waller, D.M. (2003) Direct and indirect effects of deer in for- 1943–1967. est ecosystems. Forest Ecology and Management, 181, 165–176. Levin, S.A. & Pacala, S.W. (1997) Theories of simplification on scaling of spa- Rooney, T.P., Wiegmann, S.M., Rogers, D.A. & Waller, D.M. (2004) Biotic tially distributed process. Spatial Ecology (eds D. Tilman & P. Karieva), pp. impoverishment and homogenization in unfragmented forest understorey 271–295. Princeton University Press, Princeton, New Jersey, USA. communities. Conservation Biology, 18, 787–798. Lewis, M.A. & Murray, J.D. (1993) Modeling territoriality and wolf-deer inter- Rosenzweig, M.L. (1995) Species Diversity in Space and Time. Cambridge actions. Nature, 366, 738–740. University Press, Cambridge, UK. McCaffery, K.R. (1995) History of deer populations in northern Wisconsin. Schmida, A. & Whittaker, R.H. (1981) Pattern and biological microsite effects Hemlock Ecology and Management: Proceedings of a Regional Conference in two shrub communities, southern California. Ecology, 62, 234–251. on Ecology and Management of Eastern Hemlock (eds G. Mroz & J. Mar- Schmida, A. & Wilson, M.V. (1985) Biological determinants of species diver- tin), pp. 109–114. Michigan Technological University, Houghton, Michigan, sity. Journal of Biogeography, 12,1–20. USA. Schmitz, O.J. (2006) Predators have large effects on ecosystem properties by McLaren, B.E. & Peterson, R.O. (1994) Wolves, moose and tree rings on Isle changing plant diversity, not plant biomass. Ecology, 87, 1432–1437. Royale. Science, 266, 1555–1558. Schmitz, O.J., Hamback, P.A. & Beckerman, A.P. (2000) Trophic cascades in Mech, L.D. (1977) Wolf-pack buffer zones as prey reservoirs. Science, 198, terrestrial systems: a review of the effects of carnivore removals on plants. 320–321. American Naturalist, 155, 141–153. Mech, L.D. (1994) Buffer zones of territories of gray wolves as regions of Schnitzer, S.A. & Carson, W.P. (2001) Treefall gaps and the maintenance of intraspecific strife. Journal of Mammalogy, 75, 199–202. species diversity in a tropical forest. Ecology, 82, 913–919. Mech, L.D. & Peterson, R.O. (2003) Wolf-prey relations. Wolves: Behavior, Stromayer, K.A. & Warren, R.J. (1997) Are overabundant deer herds in the Ecology, and Conservation (eds L.D. Mech & L. Boitani), pp. 131–160. Uni- eastern United States creating alternate stable states in forest plant communi- versity of Chicago Press, Chicago, Illinois, USA. ties? Wildlife Society Bulletin, 25, 227–234. Mladenoff, D.J., Sickley, T.A. & Wydeven, A.P. (1999) Predicting gray wolf Terborgh, J., Feely, K., Silman, M., Nunez, P. & Balukjian, B. (2006) Vegeta- landscape recolonization: logistic regression models vs. new field data. Eco- tion dynamics of predator-free land-bridge islands. Journal of Ecology, 94, logical Applications, 9,37–44. 253–263. Mladenoff, D.J., White, M.A., Pastor, J. & Crow, T.R. (1993) Comparing spa- Tilghman, N.G. (1989) Impacts of white-tailed deer on forest regeneration in tial pattern in unaltered old-growth and disturbed forest landscapes. Ecologi- northwestern Pennsylvania. Journal of Wildlife Management, 53, 524–532. cal Applications, 2, 294–306. USDA Forest Service (2001) CDS Data dictionary. USDA Forest Service, Mladenoff, D.J., Sickley, T.A., Haight, R.G. & Wydeven, A.P. (1995) A regio- Region 9. nal landscape analysis and prediction of favorable gray wolf habitat in the USDA Forest Service (2004) Chequamegon-Nicolet National Forest: final envi- Northern Great Lakes Region. Conservation Biology, 9, 279–294. ronmental impact statement. U.S. Department of Agriculture, Forest Service, Nelson, M.E. & Mech, L.D. (2006) A 3-decade dearth of deer (Odocoileus vir- Region 9. ginianus) in a wolf (Canis lupus)–dominated ecosystem. American Midland Van Deelen, T.R., Pregitzer, K.S. & Haufler, J.B. (1996) A comparison of pre- Naturalist, 155, 373–382. settlement and present-day forests in two Northern Michigan deer yards. van Nes, E.H. & Scheffer, M. (2005) Implications of spatial heterogeneity for American Midland Naturalist, 135, 181–194. catastrophic regime shifts in ecosystems. Ecology, 86, 1797–1807. Watkins, R.Z., Chen, J., Pickens, J. & Brosofske, K.D. (2003) Effects of forest Paine, R.T. (1980) Food webs: linkage, interaction strength and community roads on understorey plants in a managed hardwood landscape. Conservation infrastructure. Journal of Animal Ecology, 49, 667–685. Biology, 17, 411–419. Pastor, J. & Mladenoff, D.J. (1992) The southern boreal-northern hardwood White, P.J. & Garrott, R.A. (2005) Yellowstone’s ungulates after wolves – expec- border. A Systems Analysis of the Global Boreal Forest (eds H.H. Shugart, tations, realizations, and predictions. Biological Conservation, 125, 141–152. R. Leemans & G.B. Bonan), pp. 216–240. Cambridge University Press, WiDNR (1998) WISCLAND land cover (WLCGW930). Wisconsin Department Cambridge, UK. of Natural Resources, Madison, Wisconsin, USA. Peek, L.J. & Stahl, J.F. (1997) Deer management techniques employed by the WiDNR (2010) White-tailed Deer Population Status 2010. Wisconsin Depart- Columbus and Franklin County Park District, Ohio. Wildlife Society Bulletin, ment of Natural Resources, Madison, Wisconsin, USA. 25, 440–442. Wiedenhoeft, J.E. & Wydeven, A.P. (2005) Summary Report - GIS mapping of Peet, R.K., Wentworth, T.R. & White, P.S. (1998) A flexible, multipurpose historical gray wolf pack territories. Unpubl. Report. Wisconsin Department method for recording vegetation composition and structure. Castanea, 63, of Natural Resources, Park Falls, Wisconsin, USA. 262–274. Wiegmann, S.M. & Waller, D.M. (2006) Fifty years of change in northern Polis, G.A. (1999) Why are parts of the world green? Multiple factors control upland forest understories: identity and traits of “winners” and “loser” plant productivity and the distribution of biomass. Oikos, 86,3–15. species. Biological Conservation, 129, 109–123. Pregitzer, K.S. (1990) The ecology of northern white-cedar. Workshop Proceed- Wydeven, A.P., Schultz, R.N. & Thiel, R.P. (1995) Monitoring a recovering ings for the Northern White-Cedar in Michigan (ed. D.O. Lantagne), pp. 8– wolf population in Wisconsin, 1979-1991. Ecology and Conservation of 14. Michigan State University, East Lansing, Michigan, USA. Wolves in a Changing World (eds L.N. Carbyn, S.H. Fritts & D.R. Seip), Ray, J.C. (2005) Large carnivorous animals as tools for conserving biodiver- pp. 47–156. Canadian Circumpolar Institute, Occasional Publication No. 35, sity: assumptions and uncertainties. Large Carnivores and the Conservation Edmonton, Alberta, Canada. of Biodiversity (eds J.C. Ray, K.H. Radford, R.S. Steneck & J. Berger), pp. Wydeven, A.P. & Wiedenhoeft, J.E. (2010) Gray wolf population, 2009–2010. 34–56. Island Press, Washington, DC, USA. Wisconsin Wildlife Surveys, 16, 140–155. Ripple, W.J. & Beschta, R.L. (2012) Trophic cascades in Yellowstone: the first Wydeven, A.P., Wiedenhoeft, J.E., Schultz, R.N., Thiel, R.P., Jurewicz, R.L., 15 years after wolf reintroduction. Biological Conservation, 145, 205–213. Kohn, B.E. & Van Deelen, T.R. (2009) History, population growth, and Ripple, W.J., Rooney, T.P. & Beschta, R.L. (2010) Large predators, deer and management of wolves in Wisconsin. Recovery of Wolves in the Great Lakes trophic cascades in boreal and temperate ecosystems. Trophic Cascades: Pre- Region of the United States: An Endangered Species Success Story (eds A.P. dators, Prey, and the Changing Dynamics of Nature (eds J. Terborgh & J.A. Wydeven, T.R. Van Deelen & E.J. Heske), pp. 87–105. Springer, New York. Estes), pp. 141–162. Island Press, Washington, DC, USA. Rogers, D.A., Rooney, T.P., Hawbaker, T.J., Radeloff, V.C. & Waller, D.M. Received 15 October 2012; accepted 19 March 2013 (2009) Paying the extinction debt in southern Wisconsin forest understories. Handling Editor: Charles Canham Conservation Biology, 23, 1497–1506.
© 2013 The Authors. Journal of Ecology © 2013 British Ecological Society, Journal of Ecology, 101, 837–845 The Cost Of Killing Washington's Wedge Wolves . News | OPB Page 1 of 6
OPB FM ŀ Now Playing: Think Out Loud contribute (https://secure.opb.org/support/contribute/) (/)
News
Fish & Wildlife (/News/Topic/Fish-And-Wildlife/) | Ecotrope (/News/Blog/Ecotrope/) The Cost Of Killing Washington's Wedge Wolves
by Cassandra Profita (/contributor/cassandra-profita/) Ecotrope | Nov. 14, 2012 6:50 a.m. | Updated: Feb. 19, 2013 1:28 p.m.
The Spokesman Review reports
EarthFix is a public media partnership of Oregon Public Broadcasting, Idaho Public Television, KCTS9 (http://res.cloudinary.com/bdySeattle,4ger4/ima KUOW Pugetge/fetch/c_limit,h_480,w_ Sound Public Radio, 6 Northwest Public Radio and Television, Jefferson media.s3.amazonaws.com/wp- Public Radio, KLCC and the Corporation for Public content/uploads/2012/11/WAWolvesMap.jpg)Broadcasting.
In this map of Washington's wolf populations, the Wedge Pack in THANKS TO OUR SPONSORS: the northeast corner of the state is become a sponsor (/support/partners/) outlined as having been removed.
(http://www.spokesman.com/stories/2012/nov/14/repeat-of-wolf-kill- unlikely/) that the state of Washington spent an estimated $76,500 to kill a wolf pack blamed for killing or injuring 16 calves on a ranch in northeast Washington.
The Washington Department of Fish and Wildlife spent four days and about $22,000 killing six wolves in the pack using a helicopter and a marksman, the paper reports. That was after the agency spent 39 days and $54,500 killing one wolf.
Time and money weren’t the only costs, though. It also cost the agency some public support, according to spokeswoman Madonna Luers.
“Our director (Phil Anderson) has said that he never wants to do this again,” Luers said. “… The social acceptance is just not there.”
The Wedge wolves were the first to be lethally removed by WDWF (http://ecotrope.opb.org/2012/08/washingtons-first-lethal-removal- of-wolves/). The agency sent the cost estimates to the senator who chairs a legislative committee overseeing the agency. He’s planning a hearing to review the decision to kill the pack.
Wolves (/news/blog/ecotrope/tag/Wolves/)
http://www.opb.org/news/blog/ecotrope/the-cost-of-killing-washingtons-wedge-wolves/ 2/10/2016 The Cost Of Killing Washington's Wedge Wolves . News | OPB Page 2 of 6
Recent Posts More From Ecotrope (/news/blog/ecotrope/) Follow Me Over To The EarthFix Site (/news/blog/ecotrope/vestas-americas-president-were-here-to- (/news/blog/ecotrope/follow-me-over-to-the- stay/) earthfix-site/) Vestas Americas President: "We're here to stay." (/news/blog/ecotrope/vestas-americas-president-were- Underwater Camera Films Fish Habitat Off Oregon's Coast here-to-stay/) (/news/blog/ecotrope/underwater- (/news/blog/ecotrope/underwater-camera-films-fish-habitat-off-oregons- (/news/blog/ecotrope/infographic-thanksgiving-foods-grown-in- camera- coast/) oregon/) films-fish- habitat- Graphic: Thanksgiving Foods Grown In Oregon off- (/news/blog/ecotrope/infographic-thanksgiving-foods-grown-in- oregons- oregon/) coast/)
Bandon Residents "Held Hostage In Their Homes" By Mosquitoes More News (/news/blog/ecotrope/bandon- (/news/blog/ecotrope/bandon-residents-held-hostage-in-their-homes- residents- by-mosquitoes/) held- hostage- in-their- homes- by- mosquitoes/)
Q&A: Is Climate Change Making Us Sick? (/news/blog/ecotrope/is-climate- change-making-us-sick/) (/news/blog/ecotrope/is- climate- change- making- us-sick/)
Ecotrope Is Merging With EarthFix (/news/article/columbia-river-salmon-dams-tribes/) (/news/blog/ecotrope/ecotrope-is-merging-with- earthfix/) NW Tribes Seek To Restore Columbia River Salmon Runs (/news/article/columbia-river-salmon- dams-tribes/) Oregon Zoo Polar Bears Help Climate Change Researchers (/news/blog/ecotrope/oregon-zoo-polar- (/news/blog/ecotrope/oregon-bears-help-climate-change- zoo-polar- researchers/) bears- help- climate- change- researchers/)
Browse Archives by Date
Select Month
http://www.opb.org/news/blog/ecotrope/the-cost-of-killing-washingtons-wedge-wolves/ 2/10/2016 The Cost Of Killing Washington's Wedge Wolves . News | OPB Page 3 of 6
THANKS TO OUR SPONSORS
become a sponsor (/support/partners/)
(/news/article/new-klamath-dam-removal-bypass-congress/)
Could The New Klamath Dam Removal Plan Kick-Start The Stalled Water Deals? (/news/article/new-klamath-dam-removal-bypass-congress/)
(/news/series/burns-oregon-standoff-bundy-militia-news- updates/northwest-volunteers-restore-malheur-refuge-occupation- oregon/)
Northwest Volunteers Want To Help Restore Malheur Refuge (/news/series/burns-oregon- standoff-bundy-militia-news-updates/northwest-volunteers-restore-malheur-refuge-occupation- oregon/)
http://www.opb.org/news/blog/ecotrope/the-cost-of-killing-washingtons-wedge-wolves/ 2/10/2016 The Cost Of Killing Washington's Wedge Wolves . News | OPB Page 4 of 6
(/news/article/oregon-wildlife-images-golden-eagle-gets-cataract- removed-in-medford/)
Wildlife Images Golden Eagle Gets Cataract Removed In Medford (/news/article/oregon-wildlife- images-golden-eagle-gets-cataract-removed-in-medford/)
More OPB
(/artsandlife/article/jeffrey-bale-follows-his-own-path/)
Mosaic Artist Jeffrey Bale Follows His Own Path (/artsandlife/article/jeffrey-bale-follows-his-own- path/)
http://www.opb.org/news/blog/ecotrope/the-cost-of-killing-washingtons-wedge-wolves/ 2/10/2016 The Cost Of Killing Washington's Wedge Wolves . News | OPB Page 5 of 6
(/radio/programs/thinkoutloud/segment/affordable-housing-plans- preserving-yemen-history/)
Affordable Housing Plans | Defending Pete Santilli | Preserving Yemen's History (/radio/programs/thinkoutloud/segment/affordable-housing-plans-preserving-yemen-history/)
OPB | Feb. 10, 2016
About (/about/) FAQs (/about/faq/member/) Employment (/about/careers/) Volunteering (/about/volunteering/) Internships (/about/internships/) Producing for OPB (/about/produce/) Accessibility (/about/accessibility/) Hire OPB (/about/services/) Pressroom (/pressroom/) OPB Public Files (/about/#moreinfo)
Contact (/about/contactus/) Email OPB (/about/contactus/) Newsroom (/about/contactus/newsroom/) Connect (/about/connect/) Facebook (http://www.facebook.com/oregonpublicbroadcasting?_fb_noscript=1) Twitter (http://twitter.com/opb) Reception Issues (/about/contactus/reception/)
Mobile (/about/connect/apps/) Podcasts (/ondemand/#podcasts) Mobile Apps (/about/connect/apps/) Newsletters (/about/newsletter/) RSS Feeds (/about/rssfeeds/)
Support (/support/) Contribute (https://secure.opb.org/support/contribute/) Business Partners (/support/partners/) Leadership Giving (/support/membership/leadership/)
http://www.opb.org/news/blog/ecotrope/the-cost-of-killing-washingtons-wedge-wolves/ 2/10/2016 The Cost Of Killing Washington's Wedge Wolves . News | OPB Page 6 of 6
Planned Giving (/support/plannedgiving/) Vehicle Donations (/support/vehicledonation/) Volunteering (/about/volunteering/) Shop (http://www.pntrac.com/t/2-26430-54246-51107)
Shop (http://www.pntrac.com/t/2-26430-54246-51107)
Legal (/about/termsofuse/) Privacy Policy (/about/privacypolicy/) Terms of Use (/about/termsofuse/) Editorial Policy (/about/editorialpolicy/) Contest Rules (/support/drawing/)
FCC Public Files (https://stations.fcc.gov/station-profile/kopb-tv) KOPB-TV (https://stations.fcc.gov/station-profile/kopb-tv) KOAB-TV (https://stations.fcc.gov/station-profile/koab-tv) KOAC-TV (https://stations.fcc.gov/station-profile/koac-tv) KEPB-TV (https://stations.fcc.gov/station-profile/kepb-tv) KTVR (https://stations.fcc.gov/station-profile/ktvr)
Search OPB
(http://www.pbs.org/?station=KOPB) (http://www.npr.org/stations/force/force_localization.php?station=KOPB_FM&url=http://www.npr.org/)
(http://www.pri.org/) (http://www.bbc.co.uk/worldservice) (http://americanpublicmedia.publicradio.org/)
© 2016, Oregon Public Broadcasting (/)
http://www.opb.org/news/blog/ecotrope/the-cost-of-killing-washingtons-wedge-wolves/ 2/10/2016 CDFW Draft Wolf Plan Review Eisenberg
California Department of Fish and Wildlife Draft Wolf Plan
Review by Cristina Eisenberg, PhD Lead Scientist Earthwatch Institute [email protected]; [email protected]
Introduction/Overview In reviewing the Draft Wolf Plan for the State of California, I operated under the assumption that wolves would be delisted in much of the conterminous US within the next 5 years and possibly before a wolf population is established in California (p. 10, lines 23-29). This means that state protection for wolves in California under the California Endangered Species Act (CESA) may be the only form of protection these wolves will receive as their population becomes established. I have summarized my assessments by topic below, and also have made notes throughout the text of the Draft Wolf Plan using TrackChanges. The Draft Wolf Plan creates the framework for applying an adaptive management strategy to wolf conservation. This is a very wise approach. The plan reviews and does a thorough job of addressing most of the factors that can negatively impact wolves and their prey. Detail on how adaptive management would address those factors is fully described in Phase 2. However, information about how adaptive management would be applied is almost completely lacking from Phase 3. My specific concerns the plan as a whole are: 1) the need for improvement of data gathering and analysis tools (monitoring, modeling) to apply best science; 2) the low wolf population thresholds in Phases 1 and 2 do not allow for environmental stochasticity or for development of a stable wolf population; and 3) the lack of measurable population objectives in Phase 3. A fourth topic of concern not discussed in the Draft Wolf Plan pertains to coyote management that allows hunting by humans. Recent events in other states (Utah, Iowa, the northeastern US) clearly demonstrate the propensity for coyote hunters to mistake wolves for coyotes. This is leading to frequent illegal lethal take of recolonizing wolves. I strongly recommend that coyote hunting be eliminated in California, in order to enable wolf conservation to proceed. Because coyotes will have a small but significant impact on elk, I further recommend an adaptive management strategy to manage these impacts.
Background The background and literature review for the Draft Wolf Plan covers wolf ecology and ungulate ecology in depth. It clearly describes threats to wolves (with the exception of coyote hunting) and all threats to ungulate prey species typically recognized by science. It describes food web impacts, including reversal of mesopredator release, that could result from wolf recolonization, and strategies to address such impacts. The Plan also provides a thorough
1
CDFW Draft Wolf Plan Review Eisenberg
analysis of the many uncertainties with regard to what we know about the past (e.g., distribution and population of wolves), current tools (e.g., models for wolf suitable habitat), and how these uncertainties make it difficult to predict the impacts of wolves on game populations and the ultimate carrying capacity for wolves in California. The Plan makes a solid case that California is very different ecologically than any of the other states in which wolves have become established. This means that wolf recolonization and management of California’s wolf population may proceed in a very different manner in this state than elsewhere. One of the key points is that wolf carrying capacity, based on prey biomass, may be far lower in California than in places like Washington or Montana or Minnesota. The background section highlights the need for a more rigorous modeling approach to create models that will provide key tools in wolf (and ungulate) conservation decision making. Development of models developed specifically for California that utilize best science will be an important research priority (P. 14, lines 9-12).
Diseases The Draft Wolf Plan provides a comprehensive assessment of diseases that may threaten wolves. These factors are not incorporated in the strategies for wolf conservation (Table 1-A). Lessons learned from Yellowstone, where the wolf population has declined severely due to disease (Smith et al. 2014), strongly suggest acknowledging the stochastic effects of disease in establishing the wolf population levels that define each wolf conservation phase in California.
Ungulate Wolf recolonization and recovery in California will undeniably have impacts on ungulates. As the Draft Wolf Plan indicates, the strengths of these impacts are impossible to fully predict. It is expected that upon recolonization wolves will prey primarily on elk. However, given the low overall elk population in California (~11,000 individuals), and the fact that some of the individual herds are isolated and small (<500 individuals), elk may not be the primary prey of wolves initially or ever in California (pp. 106-114). Contrast this with an elk population of 150,000 in Montana, where wolves have become well established. Wolves are opportunists that primarily prey on ungulates. They take the most vulnerable animals first. They also select elk over deer where elk are abundant, because it may be more energy-efficient for them to take down an animal that is far greater in biomass than deer (p. 134, lines 32-36). Given that the California deer population is over one order of magnitude greater than the elk population, and that deer are the dominant wild ungulate by biomass in every place that wolves are likely to recolonize, I do not expect that wolf predation on elk will be as much of an issue as is predicted in the Plan. Recolonizing and established wolves have both compensatory and additive mortality effects on ungulate populations. Wolves will have additive mortality effects on prey populations that are below carrying capacity, as is the case in California. However, even in prey populations
2
CDFW Draft Wolf Plan Review Eisenberg
that are above carrying capacity, wolves can have additive mortality effects (p. 138, lines 34-37; Eisenberg et al. 2014). Therefore, any wolf conservation plan in California must address the linked matter of conservation of wolf prey species, both for predation by wolves and for utilization by humans (e.g., hunting). Without seeing the forthcoming California updated mule deer and elk management plans, I am unable to comment further on predator-prey interactions involving wolves and California’s dominant (by biomass) ungulates.
Livestock The Draft Wolf Plan does a thorough job of reviewing the history of livestock depredation in areas where wolves have become established (Washington, Oregon, the NRM). Depredation has been quite low in these areas (<1.4% of cattle and <4.2% of sheep depredation is attributable to wolves) and has been positively correlated to wolf population size. While wolf depredation on livestock will likely be similarly low in California, any depredation will create strong concern on the part of livestock producers and will need to be addressed. The Draft Wolf Plan provides a sound strategy that focuses on non-lethal deterrents to prevent and address depredation. As is pointed out in the Plan (p. 157, lines 16-26), a significant percentage of wolf packs (~28%) depredate. When properly utilized, nonlethal deterrents reduce the potential for wolf depredation. However, none will completely eliminate the potential for depredation. Wolf removal is not the answer either, because when entire packs are removed, 60% of those territories are recolonized by new packs that eventually depredate (p. 157, lines 19- 22). Addressing wolf depredation on livestock involves an adaptive management strategy that will emphasize education about and implementation of a variety of nonlethal deterrents, with financial support to livestock producers to apply such deterrents. The Draft Wolf Plan outlines such as strategy. The guidelines for lethal removal of wolves specified in Table 1-Q, Phase 2 of (4 confirmed depredation incidents by the same individual) are reasonable; however they should be continued in Phase 3. The vagueness of the proposed actions for Phase 3 (“to be determined . . . based on wolf population and legal status”) does not support wolf conservation in the long run. Lethal removal of wolves, while not a solution, may be the best approach in situations where all else has been tried.
Research/Information Management Needs Research should prioritize development of models for assessing wolf habitat suitability, wolf carrying capacity, and wolf impacts on ungulates. These models will need to be created using best science. Modeling tools developed in 2003-2006, which have been used to create models for the Draft Wolf Plan, do not constitute best science, because these models are dated. The designation of jaguar critical habitat by the US Fish and Wildlife Services, which has been problematic, provides an example of the difficulties that can arise when applying models, how
3
CDFW Draft Wolf Plan Review Eisenberg
models can be biased to produce results that reflect certain agendas, and how ineffective modeling can open the door to litigation (Menke and Hayes 2003; Hatten et al. 2005; Robinson 2005; USFWS 2013). The California Department of Fish and Wildlife (CDFW) should focus on applying best science to develop tools to conserve wolves and manage ungulates in the state. Colorado Parks and Wildlife has long had an exemplary wildlife research program. Their use of rigorous science has supported conservation strategies for carnivores and ungulates in that state that have proven defensible. California needs defensible modeling, monitoring, and management methods in place that will help achieve wolf conservation and management and stand up to litigation challenges. Development of such tools needs to be a funding priority, as litigation can create expenses far in excess of what it would take to develop these tools properly.
Conservation CESA mandates the CDFW to conserve wolves under Fish and Game Code (FGC) Section 2061. The factors that threaten wolf conservation presented in the Draft Wolf Plan, strategies that will ensure wolf conservation, and coordination with other state and federal agencies are very clearly and thoroughly described. Description of wolf habitat suitability and characteristics of said habitat are appropriate, with the exception of roads. Wolves have demonstrated their propensity to cross busy travel corridors, such as Interstate highways. The model developed by Thiel (1985) for wolves in the Midwest has not been broadly applicable to recolonizing wolves in other areas and may not apply to California. However, the map provided of potential suitable wolf habitat in California, based on factors such prey abundance, public land ownership, forest cover, human influences, and domestic sheep presence (p. 191), represents a good, albeit conservative starting point for a discussion about wolf conservation. While it is impossible for wolves to reoccupy their historic range, it is likely that they will attempt to recolonize as much of that range as is currently suitable habitat. My ability to comment more fully on how predator-prey interactions between wolves and their primary ungulate prey (by biomass) in California (deer and elk) will influence wolf recolonization and habitat suitability is hampered by the fact that I have not yet seen the forthcoming updated mule deer and elk management plans. Finally, I agree that the greater density of humans in California is likely to impact California’s potential wolf population size differently than in other regions where wolves have been assessed (p. 192, lines 36-39). Trophic cascade effects driven by wolves (a keystone predator) in California will be insignificant unless a high wolf population develops. Scientists have found that without the strong bottom-up effect of fire in an area with a high wolf population, wolves alone were unable to drive a trophic cascade in an aspen-elk system (Eisenberg 2012; Eisenberg et al. 2013; Mech 2012). Given that wolves in California may never reach a high density, as a high density of wolves may not have been historically present, and given that the dominant habitat type for ungulates is oak, trophic cascades in California, if present, will be difficult to document. Similarly, while an ecologically sustainable wolf population will be relatively straightforward to
4
CDFW Draft Wolf Plan Review Eisenberg
define, an ecologically effective one (e.g., one capable of driving a trophic cascade) will not be easily definable (Soulé et al. 2003). Thus, conserving this keystone predator as mandated by FGC section 1801, to perpetuate their intrinsic and ecological values (p. 7 lines 43-45), may be difficult to specify empirically. Nevertheless, as a keystone predator wolves are an essential component of a food web and help create greater biodiversity and ecological resiliency to climate change. Therefore, the precautionary principle indicates acknowledging the wolf’s role as a keystone predator in California, and setting up monitoring strategies to measure any ecological effects of its presence across trophic levels. The Plan does a very good job of describing potential trophic relationships. The Plan notes that wolf mortality rates are context dependent, as is the case with any other species. The Plan also reviews the complex effects of wolf mortality on wolf pack status (Borg et al. 2014). While it is unclear what effects federal wolf delisting will have on wolves in California, it is likely that these effects will be negative, creating wolf population sinks outside of California in areas that have functioned as source populations. Therefore, delisting wolves outside of California will create instability of wolf populations in California.
Strategy Overview The California wolf conservation strategy is summarized in Table 1. Below are my comments on this table. If an element is not mentioned below, that means I have no concerns or comments about it.
Element A: Planning early for next steps in wolf recovery is wise. I suggest continuing this approach, but resetting the wolf number/temporal criteria to reflect changes made to Element B , as specified below.
Element B: Phase 1 should be 5 breeding pairs for 3 consecutive years. Phase 2 should be 9 breeding pairs for 3 consecutive years. Phase 3 should specify a minimum wolf population to maintain in the state. Population levels and time in years to move to the next phase are insufficient, as currently specified in the Draft Plan, to allow for environmental stochasticity, wolf poaching, and potential need for lethal removal of wolves due to depredation.
Element L: Relocation of wolves as specified in Phase 2.3, subsequent to a reduction of allocated big game tags is not based on science, it is based on natural resources management economics. It is inappropriate to apply such an economic approach to a wolf population that is in the early stages of becoming established. It risks scapegoating wolves further, and this could have negative impacts on human perception of wolves.
Element O: Given the negligible safety threat that wolves signify for humans, I suggest striking this element.
5
CDFW Draft Wolf Plan Review Eisenberg
Element P: The allowed lethal control threshold specified in Phase 2 of total human-caused mortality not in excess of 10% is too high. Sustainable wolf take ranges widely (from 12% - 49%), as reported by various authors (Gude et al. 2001; Adams 2008; Creel and Rotella 2010). It is impossible to predict what sustainable wolf take may be in California. The precautionary principle suggests setting the threshold in Phase 2 for allowed lethal take based on human- caused mortality more conservatively—such as at 5%.
Element Q: As described for Phase 2, the approach/thresholds for lethal control of wolves CDFW proposes are sound. However, Phase 3 needs to be defined, otherwise this opens the door for lethal take without sideboards.
Element R: Lethal control of wolves to promote elk and other prey species population growth (Phases 2 and 3) is unacceptable. Other strategies need to be implemented, such as ungulate or wolf translocation. This opens the door for lethal take without sideboards and scapegoats the wolf in a system in which predator-prey relationships will be highly complex ecologically (Vucetich et al. 2004; Hebblewhite and Smith 2010).
Conclusion The Draft Wolf Plan states, “The best available scientific information is not, at this point, sufficient to predict with a high degree of confidence which habitat wolves will occupy in California, or how many wolves that habitat will support, over the long-term” (p. 7, lines 21-23). I agree with this assessment. Given the incomplete understanding of how wolf conservation may develop in California, the precautionary principle suggest a conservative approach to matters such as wolf take (e.g, avoiding take of wolves, setting robust wolf population thresholds between phases) and stopping and re-assessing the situation using best science and robust stakeholder feedback, as typified by an adaptive management strategy.
Literature Cited
Adams, L. 2008. Population characteristics and harvest dynamics of wolves in the central Brooks Range. Wildlife Monographs 170:1-25.
Borg, B. L., S. M. Brainerd, T. J. Meier, and L. R. Prugh. 2014. Impacts of breeder loss on social structure, reproduction, and population growth in a social canid. Journal of Animal Ecology 84(1):177-187.
Creel, S. and J. J. Rotella. 2010. Meta-analysis of relationships between human offtake, total mortality, and population dynamics of gray wolves (Canis lupus). PloS ONE 5(9):1-7, e12918.
Eisenberg, C., Hibbs, D. E., Ripple, W. J., & Salwasser, H. 2014. Context dependence of elk (Cervus elaphus) vigilance and wolf (Canis lupus) predation risk. Canadian Journal of Zoology, 92(8): 727-736.
6
CDFW Draft Wolf Plan Review Eisenberg
Eisenberg, C., S. T. Seager, and D. E. Hibbs. 2013. Wolf, elk, and aspen food web relationships: Context and complexity. Forest Ecology and Management 299:70-80.
Eisenberg, C. 2012. Complexity of Food Web Interactions in a Large Mammal System (Corvallis, OR: Oregon State University), dissertation, 238 pp.
Gude, J. A., M. S. Mitchell, R. E. Russell, C. A. Sime, E. E. Bangs, L. D. Mech, and R. R. Ream. 2001. Wolf population dynamics in the US Northern Rocky Mountains are affected by recruitment and human-caused mortality. Journal of Wildlife Management 76(10):108-118.
Hatten, J. R., A. Averill-Murray, and W. E. Van Pelt, W. E. 2005. A spatial model of potential jaguar habitat in Arizona. Journal of Wildlife Management 69, no. 3: 1024-33.
Hebblewhite, M., and D. W. Smith. 2010. Wolf community ecology: ecosystem effects of recovering wolves in Banff and Yellowstone national Parks. In M. Musiani, L. boitani, and P. C. Paquet, eds, The World of Wolves: New Perspectives on Ecology, Behavior, and Management (Calgary: University of Calgary Press) pp. 69-120.
Mech, D. L. 2012. Is science in danger of sanctifying the wolf? Biological Conservation 150(1): 143-149.
Menke, K. A. and Hayes, C. L. 2003. Evaluation of the Relative Suitability of Potential Jaguar Habitat in New Mexico (Santa Fe, NM: New Mexico Department of Game and Fish, 2003)
Robinson, M. 2005. Suitable Habitat for Jaguars in New Mexico (Tucson, AZ: Center for Biological Diversity).
Smith, D., D. Stahler, E. Stahler, M. Metz, K. Quimby, R. McIntyre, C. Ruhl, M. McDevitt. 2014. Yellowstone National Park Wolf Project Annual Report 2013. National Park Service, Yellowstone Center for Resources, Yellowstone National Park, Wyoming, YCR-2014-2.
Soulé, Michael E., et al. 2003. Ecological effectiveness: conservation goals for interactive species. Conservation Biology 17(5): 1238-1250.
Thiel, r. P., S. Merril, and L. D. Mech. 1998. Tolerance by denning wolves, Canis lupus, to human disturbance. Canadian Field Naturalist 12292):340-342.
Vucetich, J. A., D. W. Smioth, and Dr. R. Stauber. 2005. Influence of harvest, climate, and wolf predation on Yellowstone elk, 1961-2004. Oikos 111:259-270.
USFWS 2014. Endangered and threatened wildlife and plants; designation of critical habitat for the jaguar. Federal Register 78 (168): 53390-91.
7
The Endangered Species Act Endangered Species Handbook Page 1 of 3
Print PDF of Section or Chapter
Legislation
The Endangered Species Act
Legislation mandating a list of rare and endangered species was first enacted by Congress in 1966. The Endangered Species Act (ESA) was amended in 1969 when foreign species were added to the list. In 1973, a comprehensive model Act replaced the latter act, providing the most extensive safeguards of any legislation in the world to protect declining species.
The ESA prohibits harassing, harming, pursuing, hunting, shooting, wounding, killing, trapping, capturing and collecting listed species, unless specifically permitted, or attempting to engage in such activities within the United States or its territorial seas. Taking on the high seas is also prohibited, as are possessing, selling, delivering, carrying, transporting or shipping any species unlawfully taken within the United States, its territorial seas or on the high seas. It is also unlawful to deliver, receive, carry, transport or ship in interstate or foreign commerce in the course of commercial activity listed species, or to sell or offer listed species for sale in interstate or foreign commerce. The prohibitions apply to listed species, live and dead, their parts, and products made from their parts.
Species are listed in two categories: Endangered and Threatened. Endangered is defined as any species which is in danger of extinction throughout all--or a significant portion--of its range. Prohibitions on activities that affect species may be less strict for animals listed in the Threatened category, but these are regulated by the U.S. Fish and Wildlife Service (USFWS) on a species-by-species basis. The ESA also allows listing species similar in appearance to those that are Endangered or Threatened, when doing so would provide additional protection for the listed species.
Stiff penalties may be imposed for violations of the Endangered Species Act. Criminal activities may be punished with fines up to $50,000 and/or one year imprisonment for crimes involving endangered species, and $25,000 and/or six months imprisonment for crimes involving threatened species. Misdemeanors or civil penalties are punishable by fines up to $25,000 for crimes involving endangered species and $12,000 for crimes involving threatened species. A maximum of $1,000 can be assessed for unintentional violations. Rewards of up to $2,500 are paid for information leading to convictions.
The ESA has been extremely effective in saving wildlife and plant species in danger of extinction. Contrary to some who have claimed that the Endangered Species Act has interfered with government and private projects, there have been very few conflicts, and on the whole, these have been resolved to the satisfaction of both parties. The Northern Spotted Owl controversy was resolved by government retraining of loggers and an increase in high technology jobs that more than compensated for jobs lost. Other arrangements have been made with paper companies to protect the endangered Red-cockaded Woodpecker and, in southern California, with developers to protect the Coachella Fringe-toed Lizard. As of August 2001, the USFWS had issued 500 permits for 360 "Habitat Conservation Plans.” Many involve financial benefit to landowners. If landowners donate land where endangered species are found to a nonprofit organization or the federal government, the transaction is tax-deductible. In spite of economic interests who wish to weaken the ESA, the majority of Americans support the law and protection of endangered species (see Vanishing Species chapter).
http://www.endangeredspecieshandbook.org/legislation_endangered.php 2/10/2016 The Endangered Species Act Endangered Species Handbook Page 2 of 3
As of July 31, 2001, the Act listed 1,802 species of animals and plants as Endangered or Threatened. Of these, a total of 507 animal and plant species were native to the United States. The Act has been instrumental in saving native species, including endangered species such as the California condor, the Black-footed Ferret and the Bald Eagle, and Threatened species such as the Northern Spotted Owl. Programs of habitat protection, captive breeding, and other means of aiding in the recovery of listed species have prevented the extinction of hundreds of plant and animal species, many little known to the American public. The Hawaiian Islands have the largest number of listed species as a result of the destruction of native ecosystems and species by introduced animals and diseases, and clearing of forests for agriculture and ranching.
The Endangered Species Act has also been important in regulating the importation and exportation of exotic species listed. Foreign species listed totaled 555 animal and three plant species on July 31, 2001. These species include Leopards; Tigers; all species of rhinoceros; the great whales; the Andean Condor; Harpy Eagle; Imperial Parrot, among many other parrot species; Resplendent Quetzal; all sea turtles; numerous endangered tortoises; endangered caiman and crocodiles; iguanas; and fish, as well as seven endangered foreign invertebrates. Mammals comprise the majority of foreign species--251 Endangered and 17 Threatened (compared with only 63 Endangered and 9 Threatened U.S. species). These listings have prevented importation of many endangered species and their products. Prior to importation of any listed species, or part thereof, a permit must be obtained from the USFWS. Permits are not granted for commercial exploitation of non-captive-bred endangered species. Commercial importation of Leopard, Tiger or Ocelot skins, for example, is not allowed under the ESA. In spite of the severe penalties that can be exacted under the law, including jail sentences, illegal imports continue, and thousands are confiscated by USFWS agents each year. Many are tourist purchases, such as stuffed sea turtles, fur coats, and taxidermy specimens. Others are commercial items, such as reptile skins and, recently, products for the Asian medicine trade--Tiger bones and powdered rhinoceros horn.
For zoos, importation of live wild-caught specimens of endangered species requires ESA permits. If the species is listed on the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES), which lists many of the same species on Appendix I (the category for species threatened with extinction), both import permits from the U.S. CITES authorities and export permits from foreign CITES authorities are required. The USFWS has required rigid proof that importation of listed species would not result in declines in wild populations of that species, except for emergency situations where wild populations are under extreme threats such as uncontrolled poaching or habitat destruction. The combined burden of proof needed for ESA and CITES has been essential in preventing needless removal from the wild of endangered species. With the proliferation of wild animal parks, small zoos, and animals used in entertainment, the pressure to weaken the ESA to allow importation of endangered species has increased. One wildlife dealer wrote an editorial in a trade journal, Pet Business (March 1995), that encouraged weakening the ESA: "There is no logical reason for our government to pay hundreds of thousands of dollars, if not millions, a year to control non-indigenous endangered species." Without strict regulations under the ESA, however, the Act will lose its value in preventing commercialism of endangered species around the world.
©1983 Animal Welfare Institute
http://www.endangeredspecieshandbook.org/legislation_endangered.php 2/10/2016 The Endangered Species Act Endangered Species Handbook Page 3 of 3
http://www.endangeredspecieshandbook.org/legislation_endangered.php 2/10/2016 REVIEW
Alternative stable states occur when perturbations of sufficient magnitude and direction push ecosys- tems from one basin of attraction to another (12). Trophic Downgrading of Planet Earth Tipping points (also known as thresholds or break- points), around which abrupt changes in ecosystem James A. Estes,1* John Terborgh,2 Justin S. Brashares,3 Mary E. Power,4 Joel Berger,5 structure and function (a.k.a. phase shifts) occur, William J. Bond,6 Stephen R. Carpenter,7 Timothy E. Essington,8 Robert D. Holt,9 often characterize transitions between alternative Jeremy B. C. Jackson,10 Robert J. Marquis,11 Lauri Oksanen,12 Tarja Oksanen,13 stable states. Ecosystem phase shifts can also dis- Robert T. Paine,14 Ellen K. Pikitch,15 William J. Ripple,16 Stuart A. Sandin,10 Marten Scheffer,17 play hysteresis, a phenomenon in which the loca- Thomas W. Schoener,18 Jonathan B. Shurin,19 Anthony R. E. Sinclair,20 Michael E. Soulé,21 tions of tipping points between states differ with Risto Virtanen,22 David A. Wardle23 the directionality of change (13). A third key con- cept, connectivity, holds that ecosystems are built Until recently, large apex consumers were ubiquitous across the globe and had been for millions of years. around interaction webs within which every spe- The loss of these animals may be humankind’s most pervasive influence on nature. Although such cies potentially can influence many other species. losses are widely viewed as an ethical and aesthetic problem, recent research reveals extensive cascading Such interactions, which include both biological effects of their disappearance in marine, terrestrial, and freshwater ecosystems worldwide. This processes (e.g., predation, competition, and mutu- empirical work supports long-standing theory about the role of top-down forcing in ecosystems but also alism) and physicochemical processes (e.g., the highlights the unanticipated impacts of trophic cascadesonprocessesasdiverseasthedynamicsof nourishing or limiting influences of water, temper- disease, wildfire, carbon sequestration, invasive species, and biogeochemical cycles. These findings ature, and nutrients), link species together at an emphasize the urgent need for interdisciplinary researchtoforecasttheeffectsoftrophicdowngrading array of spatial scales (from millimeters to thou- on process, function, and resilience in global ecosystems. sands of kilometers) in a highly complex network. Taken together, these relatively simple concepts he history of life on Earth is punc- set the stage for the idea of trophic downgrading. tuated by several mass extinction events (2), during which global T 1 on April 29, 2013 biological diversity was sharply reduced. Department of Ecology and Evolutionary Biology, University of California, Santa Cruz, CA 95060, USA. 2Center for Tropical These events were followed by novel Conservation, Nicholas School of the Environment and Earth changes in the evolution of surviving Sciences, Post Office Box 90381, Duke University, Durham, NC species and the structure and function of 27708, USA. 3Department of Environmental Science, Policy, their ecosystems. Our planet is presently and Management, University of California, Berkeley, CA 94720, USA. 4Department of Integrative Biology, Valley Life Sciences in the early to middle stages of a sixth Building, University of California, Berkeley, CA 94720, USA. mass extinction (3), which, like those be- 5Division of Biological Sciences, University of Montana, fore it, will separate evolutionary winners Missoula, MT 59812, USA and Wildlife Conservation Society, 6 from losers. However, this event differs Bozeman, MT 59715, USA. Botany Department, University of Cape Town, Private Bag, Rondebosch 7701, South Africa. www.sciencemag.org from those that preceded it in two fun- 7Center for Limnology, 680 North Park Street, University of damental ways: (i) Modern extinctions are largely so doing, we demonstrate the influence of predation Wisconsin, Madison, WI 53706, USA. 8School of Aquatic and being caused by a single species, Homo sapiens, and herbivory across global ecosystems and bring Fishery Sciences, University of Washington, Box 355020, and (ii) from its onset in the late Pleistocene, the to light the far-reaching impacts of trophic down- Seattle, WA 98195, USA. 9Department of Biology, Post Office sixth mass extinction has been characterized by grading on the structure and dynamics of these Box 118525, University of Florida, Gainesville, FL 32611, USA. 10Center for Marine Biodiversity and Conservation, Scripps the loss of larger-bodied animals in general and of systems. These findings suggest that trophic down- Institution of Oceanography, 9500 Gilman Drive, La Jolla, CA apex consumers in particular (4, 5). grading acts additively and synergistically with other 92093, USA. 11Department of Biology, University of Missouri– St. Louis, One University Boulevard, St. Louis, MO 63121, USA.
The loss of apex consumers is arguably human- anthropogenic impacts on nature, such as climate Downloaded from 12 kind’s most pervasive influence on the natural and land use change, habitat loss, and pollution. Department of Biology, Section of Ecology, University of Turku, FI-20014 Turku, Finland and Department of Natural world. This is true in part because it has occurred Sciences, Finnmark University College, N-9509 Alta, Norway. globally and in part because extinctions are by their Foundations in Theory 13Department of Biology, Section of Ecology, University of very nature perpetual, whereas most other envi- Ecological theory has long predicted that major Turku, FI-20014 Turku, Finland and Department of Ecology and ronmental impacts are potentially reversible on shifts in ecosystems can follow changes in the Environmental Science, Umeå University, SE-90087 Umeå, Sweden. 14Department of Biology, Box 351800, University of decadal to millenial time scales. Recent research abundance and distribution of apex consumers Washington, Seattle, WA 98195, USA. 15School of Marine and suggests that the disappearance of these animals (9, 10). Three key elements of that theory provide Atmospheric Sciences, Stony Brook University, Stony Brook, NY reverberates further than previously anticipated the foundation for interpreting recurrent patterns 11794, USA. 16Department of Forest Ecosystems and Society, (6–8), with far-reaching effects on processes as suggestive of trophic downgrading in more re- 314 Richardson Hall, Oregon State University, Corvallis, OR 97331, USA. 17Aquatic Ecology and Water Quality Management diverse as the dynamics of disease; fire; carbon cent empirical work across ecosystems. First is the Group, Department of Environmental Sciences, Wageningen sequestration; invasive species; and biogeochem- idea that an ecosystem may be shaped by apex University, Post Office Box 8080, 6700 DD Wageningen, ical exchanges among Earth’s soil, water, and consumers, which dates back more than a century Netherlands. 18Section of Evolution andEcologyandCenterfor Population Biology, 6328 Storer Hall, University of California, atmosphere. but was popularized in the 1960s (9). This concept 19 Here, we review contemporary findings on the was later formalized as the dynamic notion of Davis, CA 95616, USA. Department of Zoology, University of “ ” British Columbia, 6270 University Boulevard, Vancouver, BC V6T consequences of removing large apex consumers trophic cascades, broadly defined as the propa- 1Z4, Canada. 20Centre for Biodiversity Research, 6270 University from nature—a process we refer to as trophic down- gation of impacts by consumers on their prey down- Boulevard, University of British Columbia, Vancouver, BC V6T grading. Specifically, we highlight the ecological ward through food webs (11). Theoretical work 1Z4, Canada. 21Post Office Box 1808, Paonia, CO 81428, USA. 22Department of Biology, University of Oulu, FI-90014 Oulu, theory that predicts trophic downgrading, consider on factors that control ecosystem state resulted 23 “ Finland. Department of Forest Ecology and Management, why these effects have been difficult to observe, and in a second key advance, the recognition of alter- Faculty of Forestry, Swedish University of Agricultural Sciences, summarize the key empirical evidence for trophic native stable states.” The topology of ecosystem SE901-83 Umeä, Sweden. downgrading, much of which has appeared in the dynamics is now understood to be nonlinear and *To whom correspondence should be addressed. E-mail: literature since the beginning of the 21st century. In convoluted, resulting in distinct basins of attraction. [email protected]
www.sciencemag.org SCIENCE VOL 333 15 JULY 2011 301 REVIEW
The loss of apex consumers reduces food chain Absent Present length, thus altering the intensity of herbivory and Consumer A the abundance and composition of plants in large- ly predictable ways (10). The transitions in ecosys- tems that characterize such changes are often abrupt, are sometimes difficult to reverse, and com- monly lead to radically different patterns and path- ways of energy and material flux and sequestration.
The Cryptic Nature of Trophic Downgrading The omnipresence of top-down control in ecosys- tems is not widely appreciated because several of Sea otter its key components are difficult to observe. The main reason for this is that species interactions, which are invisible under static or equilibrial conditions, must be perturbed if one is to witness B and describe them. Even with such perturbations, responses to the loss or addition of a species may require years or decades to become evident be- cause of the long generation times of some spe- cies. Adding to these difficulties is the fact that populations of large apex consumers have long been reduced or extirpated from much of the world. The irony of this latter situation is that we often cannot unequivocally see the effects of large apex consumers until after they have been lost Seastar from an ecosystem, at which point the capacity to on April 29, 2013 restore top-down control has also been lost. An- other difficulty is that many of the processes asso- C ciated with trophic downgrading occur on scales of tens to thousands of square kilometers, whereas most empirical studies of species interactions have been done on small or weakly motile species www.sciencemag.org Fig. 1. Landscape-level effects of trophic cascades Bass from five selected freshwater and marine ecosys- tems. (A) Shallow seafloor community at Amchitka Island (Aleutian archipelago) before (1971; photo D credit:P.K.Dayton)andafter(2009)thecollapse of sea otter populations. Sea otters enhance kelp abundance (right) by limiting herbivorous sea ur-
chins (left) (20). (B) A plot in the rocky intertidal Downloaded from zone of central California before (September 2001, right) and after (August 2003, left) seastar (Pisaster ochraceous) exclusion. Pisaster increases species diversity by preventing competitive dominance of mussels. [Photo credits: D. Hart] (C) Long Lake (Michigan) with largemouth bass present (right) and experimentally removed (left). Bass indirectly reduce phytoplankton (thereby increasing water Large reef fish clarity) by limiting smaller zooplanktivorous fishes, thus causing zooplankton to increase and phyto- E plankton to decline (26). (D) Coral reef ecosystems of uninhabited Jarvis Island (right, unfished) and neighboring Kiritimati Island (left, with an active reef fishery). Fishing alters the patterns of predation and herbivory, leading to shifted benthic dynamics, with the competitive advantage of reef-building corals and coralline algae diminished in concert with removal of large fish (66). (E) Pools in Brier Creek, a prairie margin stream in south-central Okla- homa with (right) and lacking (left) largemouth and spotted bass. The predatory bass extirpate herbiv- Bass orous minnows, promoting the growth of benthic algae (67).
302 15 JULY 2011 VOL 333 SCIENCE www.sciencemag.org REVIEW with short generation times that could be manip- tems. This is largely a consequence of natural systems reduce the abundance and production ulated at small spatial scales. Although some in- variation in food chain length (10). In some cases, of autotrophs. The largemouth bass/planktivore/ fluences of apex consumers (e.g., trophic cascades) the influence of apex consumers is to suppress zooplankton/phytoplankton system in U.S. Mid- seen in experiments scale up to systems with herbivory and to increase the abundance and pro- western lakes (26) (Fig. 1C) functions in such a larger or more mobile species (14), others are duction of autotrophs. The sea otter/kelp forest manner. harder to discern at small spatial and temporal system in the North Pacific Ocean (20) (Fig. 1A) scales (e.g., many of the indirect effects of trophic and the wolf/ungulate/forest system in temper- Effects on Ecosystem Processes cascades on ecosystem processes described be- ate and boreal North America (25) (Fig. 2C) func- Apart from small oceanic islands, all regions of low). As a result, we have an incomplete and tion in this manner. Apex consumers in other our planet supported a megafauna before the distorted picture of the influences of apex con- sumers across much of the natural world. Absent Present The Widespread Occurrence of Trophic Cascades Consumer A Despite these challenges, trophic cascades have now been documented in all of the world’smajor biomes—from the poles to the tropics and in ter- restrial, freshwater, and marine systems (table S1). Top-down forcing and trophic cascades often have striking effects on the abundance and species com- position of autotrophs, leading to regime shifts and alternative states of ecosystems (15). When the Arctic fox impacts of apex consumers are reduced or removed or when systems are examined over sufficiently large scales of space and time, their influences are B often obvious (Figs. 1 and 2). Although purpose- ful manipulations have produced the most sta- tistically robust evidence, “natural experiments” on April 29, 2013 (i.e., perturbations caused by population declines, extinctions, reintroductions, invasions, and various forms of natural resource management) corrob- orate the essential role of top-down interactions in structuring ecosystems involving species such as Jaguar killer whales (Orcinus orca)(16), lions (Panthera Cougar leo)(17), wolves (Canis lupus) and cougars (Puma concolor)(18), the great sharks (19), sea otters C (Enhydra lutris)(20), diverse mesopredators (21), www.sciencemag.org and megaherbivores (22). Although the extent and quality of evidence differs among species and systems, top-down effects over spatial scales that are amenable to experimentation have proven robust to alternative explanations (23). The impacts of trophic cascades on commu- Wolf
nities are far-reaching, yet the strength of these Downloaded from impacts will likely differ among species and ecosystems. For example, empirical research in D Serengeti, Tanzania, showed that the presence or absence of apex predators had little short-term effect on resident megaherbivores [elephant (Lox- odonta africana), hippopotamus (Hippopotamus amphibius), and rhinoceros (Diceros bicornis)] be- cause these herbivores were virtually invulnerable to predation (24). Conversely, predation accounted for nearly all mortality in smaller herbivores [oribi Wildebeest (Ourebia ourebi), Thompson’s gazelle (Eudorcas thomsonii), and impala (Aepyceros melampus)], Fig. 2. Landscape-level effects of trophic cascades from four terrestrial ecosystems. (A) Upland habitat and these species showed dramatic increases in of islands with (right) and without (left) Arctic foxes in the Aleutian archipelago. Foxes drive terrestrial abundance and distribution after the local extinction ecosystems from grasslands to tundra by limiting seabirds and thereby reducing nutrient inputs from of predators. Thus, top-down forcing in this system sea to land (47). (B) Venezuelan forests on small islands of Lago Guri (left: jaguar, cougar, and harpy is more apparent in some species than others, at least eagles absent) and mainland forest (right, predators present). A diverse herbivore guild erupted with when it is studied on relatively short time scales, the loss of predators from the island, thereby reducing plant recruitment and survival (68). (C) Riparian although the aggregate ecological impact of apex habitat near the confluence of Soda Butte Creek with the Lamar River (Yellowstone National Park) consumers here, as elsewhere, remains great (24). illustrating the stature of willow plants during suppression (left, 1997) from long-term elk browsing and Other than the inclusion of top-down forcing, their release from elk browsing (right, 2001) after wolf reintroductions of 1995 and 1996 (25). (D) there is no rule of thumb on the interplay between Decline of woody vegetation in Serengeti after eradication of rinderpest (by early 1960s) and the apex consumers and autotrophs in intact ecosys- recovery of native ungulates (by middle 1980s). Left, 1986; right, 2003 (69).
www.sciencemag.org SCIENCE VOL 333 15 JULY 2011 303 REVIEW
rise of Homo sapiens (4, 27). The apex consumers A influence their associated ecosystems through Fish abundance B Mussel growth top-down forcing and trophic cascades, which 15 in turn often lead to myriad effects on other spe- 60 cies and ecosystem processes (Figs. 3 and 4). Here, we describe some of the known or suspected 10
40 -1 indirect effects of losing these apex consumers. Herbivory and wildfire. Wildfires burn up to
mm yr 5 500 million ha of the global landscape annually, 20
consuming an estimated 8700 Tg of dry plant Catch-effort biomass, releasing roughly 4000 Tg of carbon to the atmosphere, and costing billions of dollars in 0 0 fire suppression and property loss (28). The fre- quency and extent of wildfire have been largely attributed to a warming and drying climate and Sea otters - fuel accumulation from protective wildland man- - agement practices. However, the global distribution Absent and biomass of vegetation are poorly predicted Present by temperature and rainfall (29), and recent + analyses suggest that interdependencies among predation (including disease), herbivory, plant D communities, and fire may better explain the C Gulls Bald eagles dynamics of vegetation. Such interdependencies 100 80 F I are well illustrated in East Africa, where the in- B 80 troduction of rinderpest in the late 1800s deci- 60 mated many native ungulate populations, including 60 on April 29, 2013 wildebeest (Connochaetes taurinus) and buffalo B (Syncerus caffer). Reductions of these large herbi- 40 F 40 % Diet vores caused an increase in plant biomass, which % Diet M F 20 fueled wildfires during the dry season. Rinder- 20 pest was eliminated from East Africa in the 1960s I F M through an extensive vaccination and control pro- 0 0 gram. Because of this, wildebeest and buffalo populations had recovered to what was thought Fig. 3. Trophic cascade from sea otters to sea urchins to kelp (center) has myriad effects on other species to be historically high levels by the early 1980s.
and ecological processes. The increase in kelp enhances the abundance of kelp forest fish (A)(70). www.sciencemag.org The resulting increase in herbivory drove these Enhanced kelp production increases the amount of particulate organic carbon in coastal ocean waters, systems from shrublands to grasslands, thus de- thus increasing the growth rate of filter-feeding mussels (B)(71). The presence or absence of sea otters creasing the fuel loads and reducing the frequen- influences the diet of other consumers in the coastal ecosystems (C and D). In systems with abundant sea cy and intensity of wildfires (30) (Fig. 4). Other otters, Glaucous winged-gulls (Larus glaucescens) consume mostly fish (F), whereas in systems lacking examples of the interplay between megafauna sea otters, gulls consume mostly macroinvertebrates (I) (C) (72). When sea otters were abundant in the and wildfire are the increase in fire frequency after Aleutian archipelago, bald eagles (Haliaeetus leucocephalus) fed on fish (F), mammals (M), and birds (B) the late Pleistocene/early Holocene decline of in roughly equal amounts. The loss of sea otters from this system led to a stronger reliance by the eagles on seabirds (D) (73). Blue bars from system with sea otters; brown bars from system without sea otters.
megaherbivores in Australia (31) and the north- Downloaded from eastern United States (32). Disease. The apparent rise of infectious dis- aquatic systems. The establishment of no-take ma- soils, and water—are understudied and largely eases across much of the globe is commonly rine reserves in the Channel Islands of southern unappreciated. Even so, important connections attributed to climate change, eutrophication, and California led to increases in the size and abun- between these entities have become apparent in habitat deterioration. Although these factors are dance of spiny lobsters (Panulirus interruptus)and the few instances where people have looked. undoubtedly important, links also exist between declines in population densities of sea urchins, The atmosphere. Linkages between apex con- disease and predation (33). For example, the re- which are preyed on by the lobsters. The reduced sumers and the atmosphere are known or suspected duction of lions and leopards from parts of sub- urchin densities thwarted the spread of disease in freshwater, marine, and terrestrial ecosystems. Saharan Africa has led to population outbreaks among individual sea urchins, which led to a low- Trophic cascades associated with the presence or and changes in behavior of olive baboons (Papio ered frequency of epidemics of sea urchin wasting absence of apex predatory fishes in lakes can af- anubis). The baboons, in turn, have been drawn disease within the reserves (35) (Fig. 4). In fresh- fect phytoplankton density, in turn affecting the into increasing contact with people because of their water systems, the localized rise and fall of human rate of primary production, the uptake rate of CO2, attraction to crops and other human food resources. malaria is associated with the impacts of predatory and the direction of carbon flux between lakes The increased baboon densities and their expanded fishes on planktivores, which are in turn important and the atmosphere. Where apex predatory fishes interface with human populations have led to high- consumers of mosquito larvae (36). are present in sufficient numbers, they reduce the er rates of intestinal parasites in baboons and the Physical and chemical influences. The influ- abundance of smaller planktivorous minnows, humans who live in close proximity to them (17). ences of industrialization and agriculture on thus releasing zooplankton from limitation by A similar result, involving different species and Earth’s physical environments and geochemical planktivores and increasing consumption rates processes, occurred in India, where the decline of processes are widely known. However, the con- of phytoplankton by zooplankton (Fig. 1B). This vultures also led to increased health risks from tributing effects of changes in the distribution and trophic cascade causes lakes to switch from net rabies and anthrax (34). Further examples of the abundance of apex consumers to the physical and sinks for atmospheric CO2 when predatory fishes interplay between predation and disease exist for chemical nature of our biosphere—the atmosphere, are absent to net sources of atmospheric CO2 when
304 15 JULY 2011 VOL 333 SCIENCE www.sciencemag.org REVIEW these fishes are present (37) (Fig. 4). Similar pro- control in aquatic and terrestrial systems. The ex- tional diversity) is increasingly confined to for- cesses occur in the oceans and on land. Indus- perimental exclusion of native birds from small mal protected areas. Although the establishment trial whaling during the 20th century transferred areas in Hawaii resulted in an up to 80-fold of protected areas mitigates certain threats to some 105 million tons of carbon from great increase in nonnative spider density (55) (Fig. 4). biodiversity—habitat loss and fragmentation, over- whales to the atmosphere (38), and even today Other examples include the spread of the invasive exploitation, and the spread of invasive species— whale feces return various limiting nutrients from brown tree snake (Boiga irregularis) on the oth- when large apex consumers are missing, protected the aphotic to photic zones, thereby directly en- erwise vertebrate predator–free island of Guam areas often fail to function as intended. The link hancing primary productivity (39, 40) and its in- (56), the facilitating influence of reduced fish pre- between apex consumers and species diversity fluence on carbon flux and sequestration. From dation on the invasion of zebra mussels (Dreissena can occur via a number of interaction pathways, land, the demise of Pleistocene megaherbivores polymorpha) in the Mississippi River (57), and for example, by blocking competitive exclusion may have contributed to or even largely accounted reduced abundance and spread of the introduced [predatory seastars in the rocky intertidal (59)], for the reduced atmospheric methane concentra- European green crab (Carcinus maenas)bypre- mesopredator release [coyotes (Canis latrans) tion and the resulting abrupt 9°C temperature de- dation from native blue crabs (Callinectes sapidus) maintaining small vertebrate species in chaparral cline that defines the Younger-Dryas period (41). in eastern North America (58). habitats (Fig. 4) (60)], and indirect habitat effects Soils. Leaf-eating herbivores profoundly influ- Biodiversity. Earth’s biodiversity (defined here [e.g., the loss of small vertebrates from over- ence soils and their associated biota through al- as both species diversity and the associated func- grazed and degraded riparian habitats after the tered plant allocation patterns of carbon and nutrients to the roots and rhizosphere, changing Apex the quantity and quality of litter that plants return consumer 80 to the soil. Ungulate herbivores further influence - 60 soils through trampling, compaction, and the re- Fire 40 Absent turn of dung and urine. The collective influence of - Present 20 % Burned these processes is often an effect on species com- + 0 position of the vegetation and altered successional pathways (42, 43). Predators of these herbivores 45 and the trophic cascades theysetinmotionreverse - 30 on April 29, 2013 these belowground effects (44). For example, the Disease reintroduction of wolves to Yellowstone National - 15
Park has reduced the positive indirect effects of + % Epidemics 0 ungulates on soil nitrogen mineralization and po- tentially the nitrogen supply for plant growth (45). 0 ≈0
In contrast, introduced rats (46) and arctic foxes - 2 -100 (Fig. 4) (47) have reduced soil fertility and plant Atmosphere - CO nutrition on high-latitude islands by disrupting sea- ∆ P -200 birds and their sea-to-land nutrient subsidies, with - striking effects on plant community composition. + -300 www.sciencemag.org Water. Large consumers influence the com- position and quality of both fresh and salt water 0.2 through a variety of mechanisms. For example, Soil - 0.1 the collapse of large demersal fish led to a 20%
+ % Soil P reduction in silica supply to pelagic diatoms in 0 the Baltic Sea (48). In rivers, mass spawning by 15
salmon suspends sediments, thus increasing down- Downloaded from stream sediment transport (49) (Fig. 4). This flush- 10 ing of stream bed sediments by the spawning fish Water - -3 5 and the increased circulation of fresh water through + g•m the gravel interstices of the stream bed have pos- 0 itive feedbacks on salmon populations by in- creasing oxygen for incubating eggs and fry 1000 and decreasing the frequency with which bed- Invasive 500 mobilizing floods kill salmon in these early life species - stages (50). Similarly, in terrestrial systems wolves Density 0 protect riparian trees and shrubs from over- browsing by large ungulates, in turn shading and 2 cooling the adjacent streams, reducing stream bank erosion, and providing cover for fish and Biodiversity - 1 other aquatic life (51, 52).
- Bird diversity Invasive species. A common feature of many 0 successful invasive species is that they have left Fig. 4. Examples of the indirect effects of apex consumers and top-down forcing on diverse ecosystem behind their natural predators and freed them- processes, including wildfires (30); disease (35); composition of atmosphere (37), soil (47), and fresh selves from top-down control (53). Likewise, the water (49); invadability by exotic species (55); and species diversity (60). Interaction web linkages by loss of native predators leaves ecosystems more which these processes are connected to apex consumers are shown in the center. Magnitude of effect is vulnerable to invasion by nonnative species (54). showningraphsonright.Bluebarsaredatafromsystemscontainingtheapexconsumer;brownbars There are many examples of hypersuccessful in- are data from systems lacking the apex consumer. Data replotted from original sources (cited above), vasions due to the absence or loss of top-down except raw data on native bird diversity in chaparral habitats provided by K. Crooks.
www.sciencemag.org SCIENCE VOL 333 15 JULY 2011 305 REVIEW
loss of cougars (61) or wolves and grizzly bears ecological surprises that have confronted society 30. R. M. Holdo et al., PLoS Biol. 7, e1000210 (2009). (Ursus arctos)(62) from temperate and boreal over past centuries—pandemics, population col- 31. T. F. Flannery, The Future Eaters: An Ecological History of the Australasian Lands and People (Griffin, Adelaide, forests of western North America]. lapses of species we value and eruptions of those Australia, 1994). Tree recruitment failure and the eventual trans- we do not, major shifts in ecosystem states, and 32. J. L. Gill, J. W. Williams, S. T. Jackson, K. B. Lininger, formation of forests to heaths and grasslands be- losses of diverse ecosystem services—were caused G. S. Robinson, Science 326, 1100 (2009). cause of increased ungulate herbivory illustrates or facilitated by altered top-down forcing regimes 33. R. S. Ostfeld, R. D. Holt, Front. Ecol. Environ. 2, 13 (2004). 34. A. Markandya et al., Ecol. Econ. 67, 194 (2008). the influence of large apex consumers on function- associated with the loss of native apex consumers 35. K. D. Lafferty, Ecol. Appl. 14, 1566 (2004). al diversity. This process is most clearly seen by or the introduction of exotics. Our repeated fail- 36. T. H. Mather, T. T. That, Environmental Management for contrasting areas where apex consumers have ure to predict and moderate these events results Vector Control in Rice Fields [Food and Agriculture been absent for differing lengths of time. In North not only from the complexity of nature but from Organization (FAO) Irrigation and Drainage Papers no. America, where wolves and other large carnivores fundamental misunderstandings of their root 41.1, Rome, 1984]. 37. D. E. Schindler, S. R. Carpenter, J. J. Cole, J. F. Kitchell, were not extirpated until the early 20th century, the causes. Except for controlling predators to enhance M. L. Pace, Science 277, 248 (1997). effects of their loss on plants is evident only as the fish, wild game, and livestock, resource managers 38. A. J. Pershing, L. B. Christensen, N. R. Record, G. D. recruitment failure of the younger trees. Because commonly base their actions on the assumption Sherwood, P. B. Stetson, PLoS One 5, e12444 (2010). of the longevity of adult trees, the older individ- that physical processes are the ultimate driver of 39. S. Nicol et al., Fish Fish. 11, 203 (2010). 40. J. Roman, J. J. McCarthy, PLoS One 5, e13255 (2010). uals persist in what superficially appears to be a ecological change. Bottom-up forces are ubiqui- 41. F. A. Smith, S. M. Elliott, S. K. Lyons, Nat. Geosci. 3, 374 normally functioning forest ecosystem. These ef- tous and fundamental, and they are necessary to (2010). fects are best known from various U.S. National account for the responses of ecosystems to per- 42. R. D. Bardgett, D. A. Wardle, Aboveground-Belowground Parks, where the loss of large predators a few dec- turbations, but they are not sufficient. Top-down Linkages: Biotic Interactions, Ecosystem Processes, and Global Change (Oxford Univ. Press, Oxford, 2010). ades ago has left a characteristic signal of reduced forcing must be included in conceptual overviews 43. S. A. Zimov et al., Am. Nat. 146, 765 (1995). tree growth rate (63) or recruitment failure (64)in if there is to be any real hope of understanding 44. A. E. Dunham, Oikos 117, 571 (2008). the dominant tree species. A longer time horizon and managing the workings of nature. 45. D. A. Frank, Oikos 117, 1718 (2008). 46. T. Fukami et al., Ecol. Lett. 9, 1299 (2006). can be obtained from the Canadian island of 47. D. A. Croll, J. L. Maron, J. A. Estes, E. M. Danner, G. V. Byrd, Anticosti, where white-tailed deer (Odocoileus References and Notes Science 307, 1959 (2005). 48. T. Katz et al., Global Biogeochem. Cycles 23, GB4032 virginianus) have persisted in the absence of pred- 1. D. H. Janzen, Nat. Hist. 83, 48 (1974). (2009). ators for more than a century, causing the suc- 2. D. M. Raup, J. J. Sepkoski Jr., Science 215, 1501 (1982). 49. J. W. Moore et al., Ecology 88, 1278 (2007). on April 29, 2013 3. D. B. Wake, V. T. Vredenburg, Proc. Natl. Acad. Sci. U.S.A. cessive elimination of saplings of less and less 50. D. R. Montgomery, J. M. Buffington, N. P. Peterson, 105 (suppl. 1), 11466 (2008). palatable trees and shrubs and increasing gram- D. Schuett-Hames, T. P. Quinn, Can. J. Fish. Aquat. Sci. 4. F. A. Smith et al., Ecology 84, 3403 (2003). inoid dominance in the understory (65). The 53, 1061 (1996). 5. J. Ray, K. H. Redford, R. Steneck, J. Berger, Eds., Large 51. R. J. Naiman, K. H. Rogers, Bioscience 47, 521 (1997). Scottish island of Rùm, from which wolves have Carnivores and the Conservation of Biodiversity (Island, 52. R. L. Beschta, W. J. Ripple, Earth Surf. Process. Landf. 31, been absent for 250 to 500 years, provides a view Washington, DC, 2005). 1525 (2006). of the likely final outcome of predator loss and 6. E. T. Duffy, Ecol. Lett. 6, 680 (2003). 53. C. S. Kolar, D. M. Lodge, Trends Ecol. Evol. 16, 199 (2001). 7. D. R. Strong, K. T. Frank, Annu. Rev. Environ. Resour. 35, elevated herbivory in many temperate forests. 54. N. O. L. Carlsson, O. Sarnelle, D. L. Strayer, Front. Ecol. 1 (2010). Environ. 7, 525 (2009). Rùm has transitioned over this same period from 8. O. J. Schmitz, D. Hawlena, G. C. Trussell, Ecol. Lett. 13, 55. D. S. Gruner, Ecology 85, 3010 (2004). a forested environment to a treeless island. 1199 (2010). 56. J. A. Savidge, Ecology 68, 660 (1987). www.sciencemag.org These examples support the conclusion that 9. N. G. Hairston, F. E. Smith, L. B. Slobodkin, Am. Nat. 94, 57. M. R. Bartsch, L. A. Bartsch, S. Gutreuter, J. N. Am. 421 (1960). disruptions of trophic cascades due to the decline Benthol. Soc. 24, 168 (2005). 10. S. D. Fretwell, Oikos 50, 291 (1987). 58. C. E. deRivera, G. M. Ruiz, A. H. Hines, P. Jivoff, Ecology of predation constitute a threat to biodiversity from 11. R. T. Paine, J. Anim. Ecol. 49, 667 (1980). 86, 3364 (2005). within for which the best management solution is 12. R. C. Lewontin, Brookhaven Symp. Biol. 22, 13 (1969). 59. R. T. Paine, Am. Nat. 103, 91 (1969). likely the restoration of effective predation regimes. 13. M. Scheffer, S. Carpenter, J. A. Foley, C. Folke, B. Walker, 60. K. R. Crooks, M. E. Soulé, Nature 400, 563 (1999). Nature 413, 591 (2001). 61.W.J.Ripple,R.L.Beschta,Biol. Conserv. 133, 397 (2006). A Paradigm Shift in Ecology 14. O. J. Schmitz, Oecologia 145, 224 (2005). 62. J. Berger, P. D. Stacey, L. Bellis, M. P. Johnson, 15. J. Terborgh, J. A. Estes, Eds., Trophic Cascades: Predators, Ecol. Appl. 11, 947 (2001). Prey, and the Changing Dynamics of Nature (Island, The accumulation of theoretical and empirical evi- 63. B. E. McLaren, R. O. Peterson, Science 266, 1555 (1994). Downloaded from dence calls for an altered perspective on top-down Washington, DC, 2010). 64. R. L. Beschta, W. J. Ripple, Biol. Conserv. 142, 2401 (2009). forcing in ecosystem dynamics. Many practicing 16. J. A. Estes, M. T. Tinker, T. M. Williams, D. F. Doak, 65. C. Cabascon, D. Pothier, For. Ecol. Manage. 253, 112 Science 282, 473 (1998). ecologists still view large animals in general, and (2007). 17. J. Brashares, L. R. Prugh, C. J. Stoner, C. W. Epps, in (15), 66. S. A. Sandin et al., PLoS One 3, e1548 (2008). – apex consumers in particular, as ecological pas- pp. 221 240. 67. M. E. Power, W. J. Matthews, A. J. Stewart, Ecology 66, – sengers riding atop the trophic pyramid but having 18. W.J.Ripple,T.P.Rooney,R.L.Beschta,in(15), pp. 141 162. 1448 (1985). little impact on the structure below. The influences 19. R. A. Myers, J. K. Baum, T. D. Shepherd, S. P. Powers, 68. J. Terborgh et al., Science 294, 1923 (2001). C. H. Peterson, Science 315, 1846 (2007). of these animals, although acknowledged in par- 69. A. R. E. Sinclair et al., Conserv. Biol. 21, 580 (2007). 20. J. A. Estes, D. O. Duggins, Ecol. Monogr. 65, 75 (1995). 70. S. E. Reisewitz, J. A. Estes, C. A. Simenstad, Oecologia ticular cases, are generally regarded as anomalous, 21. L. R. Prugh et al., Bioscience 59, 779 (2009). 146, 623 (2006). occurring in some systems but not in many others. 22. R. N. Owen-Smith, Megaherbivores: The Influence of Very 71. D. O. Duggins, C. A. Simenstad, J. A. Estes, Science 245, This perception has generally led to the require- Large Body Size on Ecology (Cambridge Univ. Press, 170 (1989). ment of independent study and confirmation for Cambridge, 1992). 72. D. B. Irons, R. G. Anthony, J. A. Estes, Ecology 67, 1460 23. O. J. Schmitz, P. A. Hambäck, A. P. Beckerman, Am. Nat. (1986). each species and system before the null hypothesis 155, 141 (2000). 73. R. G. Anthony, J. A. Estes, M. A. Ricca, A. K. Miles, that they serve no important ecological role can 24. A. R. E. Sinclair, S. Mduma, J. S. Brashares, Nature 245, E. D. Forsman, Ecology 89, 2725 (2008). be rejected. We argue that the burden of proof be 288 (2003). Acknowledgments: Financial and logistic support provided shifted to show, for any ecosystem, that consumers 25. W. J. Ripple, R. L. Beschta, For. Ecol. Manage. 184, 299 by the Institute for Ocean Conservation Science, Defenders (2003). of Wildlife, and the White Oak Plantation; research support do (or did) not exert strong cascading effects. 26. S. R. Carpenter et al., Ecol. Monogr. 71, 163 (2001). by the NSF (various U.S. authors), Natural Sciences and 27. F. A. Smith et al., Science 330, 1216 (2010). Engineering Research Council of Canada (A.R.E.S.), Conclusions 28.J.S.Levine,T.Bobbe,T.Ray,R.G.Witt,N.Singh,Wildland and Nordforsk (L.O., T.O., R.V.). C. Eisenberg, G. Mittelbach, Unanticipated changes in the distribution and Fires and the Environment:A Global Synthesis [United J. Moore, D. Reznick, O. Schmitz, and T. Williams Nations Environment Programme Division of Environmental commented on the manuscript, and E. Smith assisted in abundance of key species have often been at- Information, Assessment, and Early Warning Technical manuscript preparation. tributed in some unspecified manner to the “com- Report (UNEP/DEIAEW/TR) 99.1, Nairobi, Kenya, 1999]. plexity of nature.” We propose that many of the 29. W. Bond, in (15), pp. 275–286. 10.1126/science.1205106
306 15 JULY 2011 VOL 333 SCIENCE www.sciencemag.org
Executive Summary
Approximately 229 million acres of federal public lands in the western United States are used for livestock grazing for cattle and sheep. The Bureau of Land Management (BLM) and the United States Forest Service (USFS) are the two federal agencies with by far the largest grazing programs among federal agencies. These programs exist mostly on the grasslands, deserts, sagebrush steppe and national forests.
Each year in January, the federal government establishes the fee it charges livestock operators to use federal public lands for grazing privileges.
In advance of the release by the Bureau of Land Management of the 2015 federal grazing fee, we have prepared this report that focuses on the extent of the federal grazing program on BLM and USFS lands and associated appropriations and receipts from grazing fees, which are an indication of the cost to the taxpayer. This report is an update of an earlier 2002 study, Assessing The Full Costs of the Federal Grazing Program.
Key Findings
1. Receipts from grazing fees were $125 million less than federal appropriations in 2014. Total federal appropriations for the USFS and BLM grazing programs in fiscal year 2014 were $143.6 million, while grazing receipts were only $18.5 million.
Appropriations for the BLM and USFS grazing programs have exceeded grazing receipts by at least $120 million annually since 2002. Had the federal government charged the average private forage market rate for non-irrigated lands in the western states, grazing receipts would have been on average $261 million, greatly exceeding annual appropriations.
2. The gap between federal grazing fees and private land fees has widened considerably. The federal grazing fee in 2014 was set at the legal minimum of $1.35/AUM, or animal unit month, which is the amount of forage to feed a cow and calf for one month. The annual federal grazing fee has been set at the minimum required by law since 2007.
In 2013, the federal grazing fees of $1.35/AUM were just 6.72 percent of fees charged for non- irrigated private grazing lands in the West, which averaged $20.10 per AUM. The gap has widened considerably since 1981, when the federal fee was 23.79 percent of fees charged on private rangelands. The federal grazing fee is generally also considerably lower than fees charged on state-owned public lands.
3. The federal grazing subsidy is even larger when all costs to the taxpayer are accounted for. Indirect costs for livestock grazing include portions of different federal agencies budgets, such as the USDA Wildlife Services, which expends money to kill thousands of native carnivores each year that may threaten livestock; U.S. Fish and Wildlife Service, which expends part of its budget for listing species as threatened or endangered resulting from harm by livestock grazing; and other federal land management agencies that expend money on wildfire suppression caused by invasive cheat grass that is facilitated by livestock grazing. The full cost of the federal grazing program is long overdue for a complete analysis.
1 Table of Contents
Executive Summary 2
Abbreviations 4
List of Tables 5
List of Figures 5
About the Authors 6 Introduction
Chapters
1. Public Lands Ranching – A Brief History 8
2. Extent of the Federal Lands Livestock Grazing Program 10 a. Acres of BLM and USFS Western Lands b. Acres of BLM and USFS Western Grazing Lands c. BLM and USFS - Animal Unit Months, Permits and Leases
3. Cost of the Federal Livestock Grazing Program 16 a. Grazing Receipts and their Distribution b. Grazing Appropriations c. Difference Between Appropriations and Receipts d. PRIA Fee Impacts on Receipts and Scenarios for Reducing Taxpayer Costs
4. BLM and USFS Grazing Fees 20 a. History of Fee Setting Approaches b. PRIA Formula Explained
5. PRIA Fees Compared to Private, State and Other Federal Grazing Fees 25 a. PRIA Fees Compared to Private Fees b. Fees Charged by Other Federal Agencies c. Fees Charged by State Agencies
6. Indirect Costs of Federal Public Lands Grazing 30
Conclusions 31
Appendices A. Indirect Costs of federal Lands Grazing 33 B. Detailed Data Tables 37
Bibliography 41
2 Abbreviations
AUM Animal Unit Month BCPI Beef Cattle Price Index BLM Bureau of Land Management CF Calculated Fee
CFR Code of Federal Regulations
CRS Congressional Research Service DOD Department of Defense DOE Department of Energy DOI Department of the Interior FLPMA Federal Land Policy and Management Act FN Footnote
FVI Forage Value Index
FWS U.S. Fish and Wildlife Service
FY Fiscal Year
GAO General Accounting Office
NASS National Agricultural Statistics Service
NFS National Forest System
NPS National Park Service
PPI Prices Paid Index
PRIA Public Rangeland Improvement Act
RBF Range Betterment Fund
USDA United States Department of Agriculture
USDOI United States Department of the Interior
USFS United States Forest Service
3 List of Tables
Table 1: BLM and USFS Total Acreage in 11 Western States, 2010 Table 2: BLM Authorizations of Permits/ Leases and AUMs, 2002-2013 Table 3: USFS Authorized Number of Permittees and AUM, 2002-2013 Table 4: USFS and BLM – Permit, Leases, Permittees and AUM for Grazing Seasons, 2002- 2013 Table 5: USFS and BLM Grazing Receipts 2002 to 2014, in 2014 Dollars Table 6: Distribution of Fee Receipts by Agency and Land Classification Table 7: BLM and USFS Direct Grazing Appropriations 2002 to 2014, in 2014 Dollars Table 8: Total BLM and USFS Direct Grazing Appropriations vs. Receipts 2002-2014, in 2014 Dollars Table 9: PRIA-Based Grazing Fees from 1981 to 2014 Table 10: PRIA Fee Calculation 1980 through 2014 and Comparison to Private Rates Table 11: Private Rates and PRIA Fees per AUM for Non-Irrigated Land in 16 Western States, 2002 and 2013 Table 12: Fees Charged by Other Federal Agencies in 2004 Table 13: Grazing Fees Charged by State Land Agencies in Western States in 2004 and 2010
Appendix A
Table A1: BLM Budget Items Potentially Containing Indirect Costs of Grazing Table A2: USFS Budget Items Potentially Containing Indirect Costs of Grazing Program Table A3: Other Federal Agencies’ Indirect Costs of Grazing on Federal Lands Table A4: Indirect Costs on State and Local Level
Appendix B
Table B1: BLM Acres Grazed by State, 2004 Table B2: USFS Acres Grazed by State, 2004 Table B3: USFS Grazing Receipts and Appropriations 2002 to 2014, in 2014 Dollars Table B4: BLM Grazing Receipts and Appropriations 2002-2004, in 2014 Dollars Table B5: Scenario of USFS and BLM Grazing Revenues with Application of Private Grazing Rates 2002-2012 Table B6: PRIA Fees and Private Fees 1981 to 2014 (Nominal and Adjusted for Inflation)
List of Figures
Figure 1: BLM Acres Grazed, 2004 Figure 2: USFS Regions Figure 3: USFS Acres Grazed, 2004 Figure 4: Annual Federal Grazing Appropriations, Receipts and Taxpayer Costs: 2002-2014 Figure 5: Annual Grazing Fees for BLM and USFS and Private Land: 1981-2013 Figure 6: The PRIA Fee As a Percentage of Private Rates: 1980-2012
4 This report was prepared for the Center for Biological Diversity.
About the Authors
Christine Glaser received her Masters degree in Economics from University of Mainz, Germany, in 1981 and completed a doctorate (rerum politicarum) at the University of Munich in 1984. She has been a partner with GreenFire Consulting Group, LLC since 2003.
Chuck Romaniello received his MS in Agricultural Economics from the University of Arizona. He worked as a natural resource/agricultural economist at the University of Arizona, Department of Agriculture, and Department of the Interior, a career spanning 32 years. He is currently retired from federal service and is living in the mountains of southwestern Colorado.
Karyn Moskowitz received her MBA in Environmental Management from the University of Washington Foster School of Business in 1995. She has been a partner with GreenFire Consulting Group, LLC since 2003.
5 Introduction
Several federal agencies permit livestock grazing on public lands in the United States, the largest being the U.S. Department of the Interior’s Bureau of Land Management (BLM) and the Department of Agriculture’s United States Forest Service (USFS). The vast majority of livestock grazing on BLM and USFS rangelands occurs in the 11 western states of Arizona, California, Colorado, Idaho, Oregon, Montana, New Mexico, Nevada, Utah, Washington and Wyoming. Rangelands are non-irrigated and generally have vegetation that consists mostly of grasses, herbs and/or shrubs. They are different from pastureland, which may periodically be planted, fertilized, mowed or irrigated.
In 2013, the BLM issued 15,739 permits to livestock operators and there were 5,711 livestock operators who had permits to graze in the national forest system. The numbers of USFS and BLM permits and livestock permittees are not directly additive, but due to a number of livestock operators who have permits from both agencies and/or more than one grazing permit per agency, the total number of livestock operators is likely to be fewer than 21,540. This compares to the approximately 800,000 ranchers and cattle producers in the United States. (Statistic Brain, 2012). The number of operators benefitting from the USFS and BLM grazing program in the West is therefore less than 2.7 percent of the nation’s total livestock operators.
By statutory direction, BLM and USFS are supposed to manage their lands for multiple purposes. Some of the major extractive uses that these lands may be subject to are grazing, mining, logging, and energy (oil, gas, coal, geothermal, wind and solar) development. To repair past and current ongoing damage, the agencies have programs to restore ecosystems, protect wildlife and wildlife habitat, and water and soil resources. As one of the major multiple uses, grazing occurs on large expanses of BLM and USFS acreage in the West.
Other federal agencies with a land base, including the National Park Service, Fish and Wildlife Service, Department of Defense (Army, Army Corps of Engineers, Air Force and Navy), and Bureau of Reclamation, often permit some livestock grazing on their lands as well, but their landholdings are small in comparison. The Bureau of Indian Affairs tribal lands and state trust lands are the other major categories of non-public lands on which livestock grazing is permitted.
The focus of this report is on federal public lands managed by BLM and USFS, and the direct costs of these federal grazing programs to taxpayers. We examine the history of public lands ranching and how much grazing the BLM and USFS agencies have authorized between 2002 and 2013. A major portion of this report deals with BLM and USFS grazing fees. We explore in detail the fee formula that has been applied since the passage of the 1978 Public Rangelands Improvement Act (PRIA). We compare the PRIA fees to rates charged on private lands, state lands, and lands managed by other federal agencies.
Appendix A explores some of the indirect costs resulting from federal grazing programs.
6 1. Public Lands Ranching - A Brief History
After the U.S. Civil War, the western range livestock industry started to expand. Access to federal lands was not regulated or restricted. Nomadic herders drove large herds of livestock from place to place in search of forage, or herds were left roaming unattended and only rounded up for branding and marketing, or to move them between summer and winter ranges. However, some ranchers with homesteads let cattle roam only during the summer, and brought them close to the home ranch in the winter to feed them with hay. (BLM Utah)
Faced with overuse and land degradation – the classic tragedy of the commons — by the 1880s ranchers developed ways to protect what they considered their customary range by using barbed wire to fence in large areas of public land, or by finding ways to monopolize water sources, access to which is a necessity for maintaining livestock herds in the arid West. (BLM Utah)
During the 1890s, severe land degradation (from grazing as well as logging) led Congress to take steps to create “forest reserves.” In 1905, the newly established Forest Service renamed these lands “National Forests,” with the Forest Service being given authority to “permit, regulate, or prohibit grazing in the forest reserves.” This did not immediately lead to a reduction of overuse. Full grazing privileges were given to livestock owners in the beginning, and stockmen were to be given “ample opportunity to adjust their business to the new conditions.” (USDA 1905, 20) One immediate consequence of the regulations was the elimination of nomadic herders from the lands managed by the Forest Service. (International Society for the Protection of Mustangs and Burros)
In the western states outside of the national forests, access to vast areas of federal lands remained open, and they continued to deteriorate until passage of the Taylor Grazing Act of 1934. The Act directed the secretary of the Interior “to stop injury to the public grazing lands by preventing overgrazing.” A newly established Division of Grazing (renamed the Grazing Service in 1939) delineated allotments, issued grazing permits and collected fees. After enactment of the Taylor Grazing Act, livestock numbers were significantly reduced and nomadic sheep and cattle herding was eliminated. (BLM Utah; BLM Wyoming)
During the Great Depression, under the authority of the Bankhead-Jones Farm Tenant Act of 1937, the federal government recovered millions of acres of failed western homesteads. Many of these lands in California, Montana, New Mexico and Texas were transferred to the Grazing Service or General Land Office. Other large parcels came under Forest Service management and were later named “national grasslands.” (Olson 1997, 4-7)
In 1946, the Grazing Service and General Land Office were merged to form the Bureau of Land Management (BLM), giving the BLM jurisdiction not just over rangelands, but also over public land minerals and land transfers, among others. (Gorte 2012, 10)
In the 1960s and 1970s, federal laws were passed to protect the environment, among them the National Environmental Policy Act (NEPA) of 1969 and the Endangered Species Act (ESA) of 1973. These laws changed the expectation as to how public lands were to be managed, leading, among other things, to changes in the terms and conditions that applied to grazing leases and permits. (BLM 3/28/14)
7 In 1974, the Forest and Rangeland Renewable Resources Planning Act was passed—and later amended in 1976 as the National Forest Management Act. This act called for management of renewable resources on national forest lands through “analysis of environmental and economic impacts, coordination of multiple use and sustained yield opportunities as provided in the Multiple-Use Sustained-Yield Act of 1960.” (74 Stat. 215; 16 U.S.C. 528–531) It also called for public participation in the development of the program.” (USDA History of Forest Planning)
In 1976, the Federal Land Policy and Management Act (FLPMA) passed into law, giving the BLM the mission to manage public lands for multiple uses and “sustained yield.” (USDOI, 2001, 1) FLPMA also established a Range Betterment Fund into which half of all BLM and USFS grazing fee receipts were to be directed for range improvements. (BLM WYOMING)
In 1978 the Public Rangelands Improvement Act (PRIA) was passed, which provided a formula for setting grazing fees on both Forest Service and BLM lands in 16 western states.1 After a trial period of seven years, this formula was made permanent by Executive Order 12548 (Feb. 14, 1986). Executive Order 12548 established a fee minimum of $1.35 per Animal Unit Month (AUM)2, and provided that annual fee adjustment could not exceed 25 percent of the previous year’s fee. (Vincent 2012, 3)
The application of the PRIA fee formula has ultimately led to BLM and USFS grazing fees that increasingly diverge from rates charged by private landowners as well as other federal and state agencies.
Under PRIA, both the BLM and USFS divide their rangelands into allotments. Allotments can vary in size from a few acres to hundreds of thousands of acres of land, and may be intermingled with private lands. BLM uses a grazing permit system to “permit” grazing allotments within its grazing districts. Outside grazing districts, BLM leases its fragmentary lands to ranchers. The Forest Service does not have grazing districts, and uses permits to authorize grazing within its allotments. (GAO 2005, 12)
To be eligible for a permit or lease on one of BLM’s allotments, livestock operators are required to own or control private “base property” that can serve to support the livestock with water, or necessary feed. Under USFS guidance, permits are issued to purchasers of permitted livestock or base property. (GAO 2005, 12)
Grazing is administered primarily through issuance of 10-year term permits for discrete grazing allotments. The 10-year permits can be renewed without competition. The current permittee or lessee has priority over others who may be interested in the permit or lease and thus has what is known as “preference.” Permittees do not obtain title to federal lands through their grazing permits and leases. (GAO 2005, 16)
1 These 16 states are: Arizona, California, Colorado, Idaho, Kansas, Montana, Nebraska, Nevada, New Mexico, North Dakota, Oklahoma, Oregon, South Dakota, Utah, Washington, and Wyoming. See Vincent 2012, 1, listing authorities that govern grazing on BLM and USFS lands in 16 contiguous Western states as P.L. 95-514, 92 Stat. 1803; 43 U.S.C. §§1901, 1905. Executive Order 12548, 51 Fed. Reg. 5985 (February 19, 1986). 2 Animal Unit Month (AUM) is a “standardized unit of measurement of the amount of forage necessary for the complete sustenance of one animal unit for a period of 1 month.” (BLM PLS 2012, 255). More precisely, it is “the use of public lands by one cow and her calf, one horse, or five sheep or goats for a month.” (BLM 1/31/14)
8 2. Extent of the Federal Lands Livestock Grazing Program
a. Acres of BLM and USFS Western Lands
As reported by Gorte et al. (2012, 19) the BLM administers 174.5 million acres of federal lands outside of Alaska, and these are almost exclusively concentrated (99.78 percent) in 11 western states (Table 1). The 11 western states are Arizona, California, Colorado, Idaho, Oregon, Montana, New Mexico, Nevada, Utah, Washington and Wyoming. Gorte et al. also report that USFS administers 141.7 million acres in these 11 western states, or 82.94 percent of its total federal land holdings. (Table 2) The delineation of the “West” in Gorte et al. is different from the delineation of the “West” with regard to PRIA, which includes 16 western states. The additional states are Kansas, North Dakota, South Dakota, Oklahoma and Nebraska.
Table 1: BLM and USFS Total Acreage in 11 Western States, 2010
Total Total Land BLM and USFS % Holdings in Land Holdings Holdings Other Total Western States 11 Western States States (Outside Alaska) (acres) (Outside Alaska)* (acres) (acres) USFS 141,762,880 29,161,710 170,924,590 82.94% BLM 174,512,265 388,054 174,900,319 99.78% Total BLM & USFS 316,275,145 29,549,764 345,824,909 91.46% *BLM and USFS lands in Alaska are not subject to PRIA or the Taylor Grazing Act and are therefore not included in this analysis. (GAO 2005, 4, 15, 55; BLM Public Land Statistics 2013, 255; Vincent 2012, 1). Sources: Gorte et al. 2012, 19
b. Acres of BLM and USFS Western Grazing Lands
Western BLM Lands Grazed
Livestock grazing is the prevailing use of BLM lands, with 137.7 million acres, or 79 percent out of 174.5 million acres of BLM land in the West authorized for livestock grazing in 2004. (GAO 2005, 15) Acres grazed differ from year to year, and were especially low in 2004 because of the drought. (GAO 2015, 14)
Figure 1 (Table B1) shows acres authorized for grazing for each of the 11 western states where BLM holds land. The four states with the largest holdings of BLM grazing lands are Nevada, Utah, Wyoming and New Mexico.
Western USFS Acres Grazed
The USFS is organized by regions, not by states. The following map (Figure 2) delineates the USFS regions and Figure 3 (Table B2) shows how many acres of grazing land the USFS administers in each of its regions. Grazing lands administered by the USFS include national forest lands and national grasslands, combined they comprise the national forest system.
9
Figure 1
Ninety-nine percent (92.1 million out of 92.9 million acres) of all USFS grazing on national forest and grasslands are in the western Regions within 16 contiguous western states: Arizona, California, Colorado, Idaho, Kansas, Montana, Nebraska, Nevada, New Mexico, North Dakota, Oklahoma, Oregon, South Dakota, Utah, Washington and Wyoming. (USDA Forest Service, Grazing Statistical Summary FY 2013, 96-97) The majority of national grasslands are located in the Great Plains states of Colorado, North Dakota, South Dakota and Wyoming. (Olson, 1997, 3)
The regions with the largest extent of USFS grazing lands are the Intermountain region (Nevada, Utah and Idaho), the Southwest region (Arizona and New Mexico), and the Rocky Mountain region (Colorado, South Dakota, Kansas, Nebraska and Wyoming).
10 Figure 2: Forest Service Regions
Source: USDA Grazing Statistical Summery 2013, 1.
Figure 3
11 Other Federal Grazing Lands
In addition to BLM and USFS, other federal agencies allow grazing on their lands, including National Park Service, Bureau of Reclamation, U.S. Fish and Wildlife Service, Department of Energy, and Department of Defense. Together, they only manage a total of 4 million acres of grazing land nationwide. (GAO 2005, 17) c. BLM and USFS—Animal Unit Months, Permits and Leases
The Animal Unit Month (AUM) is the measure of forage—plants that are eaten by livestock— used by federal land management agencies to allocate land for grazing. An AUM is defined as a “standardized unit of measurement of the amount of forage necessary for the complete sustenance of one animal unit for a period of 1 month.” (BLM Public Land Statistics 2012, 255) More precisely, it is “the use of public lands by one cow and her calf, one horse, or five sheep or goats for a month.” (BLM 1/31/14) The land area needed to produce an AUM will differ considerably depending on soil productivity and precipitation.
BLM Permits and Leases
BLM divides its rangelands into allotments. Allotments can vary in size from a few acres to thousands of acres of land, and may be intermingled with private lands. Grazing on BLM lands requires a Section 3 permit or a Section 15 lease. Section 3 permits are issued for allotments on public lands within the grazing districts. Fragmented BLM lands outside of grazing districts are known as Section 15 leases. (GAO 2005, 12) Permits and leases set out terms and conditions for grazing on BLM-managed lands and specify forage use (AUMs), season of use, and length of season. They generally cover a 10-year period and are renewable if the BLM determines that the terms and conditions of the expiring permit or lease have been met. (BLM 3/28/14) Both permits and leases specify the number of AUMs that a rancher may graze on a particular piece of land. Table 2 shows the number of BLM permit and lease authorizations issued from FY 2002 to FY 2013 as well as the associated authorized AUMs.
For FY 2013, BLM authorized 15,739 permits and leases, with a total of 8,513,271 AUMs. Table 2 shows that between FY 2002 and 2013, BLM authorized an average of 15,870 permits and leases annually, and an average of 8,359,496 AUMs. The lowest number of AUMs was 7,493,419 in 2003 and the highest was 8,985,228 in 2011.
USFS Permits
There are two notable differences between BLM and USFS reporting. Where BLM reports the “number of permits,” USFS reports the “number of permittees.” A permittee can hold multiple permits.
Further, USFS reports grazing use for the grazing season, and not for the fiscal year. BLM reports both, but did not have numbers for the 2013 grazing season. Therefore, above, we used the BLM fiscal year data, for USFS, we will use grazing season data.
12 Table 2: BLM Authorizations of Permits/ Leases and AUMs, 2002-2013
Fiscal Year Total Number of Total AUMs Section 3 Permits and Authorized Section 15 Leases (Section 3 and Authorized Section 15)
15,851 8,287,394 2002 2003 15,472 7,493,419
2004 15,544 7,574,004
2005 15,462 7,816,949 2006 16,416 8,515,292
15,587 8,546,253 2007 2008 17,292 8,531,813
2009 15,612 8,594,912
2010 15,751 8,673,822 2011 15,897 8,985,228
15,815 8,781,600 2012 2013 15,739 8,513,271
Average 15,870 8,359,496
Source: BLM Public Land Statistics from FY 2003 to FY 2013, Tables 3-7a and b, and 3-8a and b.
There were a total of 5,711 livestock operators who received permits on national forest service land during the 2013 grazing season, for 6,388,964 AUMs. Table 3 shows that between 2002 and 2013 USFS authorized permits for an average of 5,940 permittees, and an average of 6,335,542 million AUMs. The lowest number of AUMs authorized was 5,288,091 in 2004, the highest was 7,056,298 in 2010.
BLM and USFS Combined Totals
Table 4 shows the number of BLM permits, USFS permittees, and associated total AUMs for the grazing season. BLM numbers differ somewhat from Table 2 because they are reported by grazing season rather than for the fiscal year, to match the USFS’s reporting. Because BLM’s Public Lands Statistics did not report 2013 grazing season numbers, there are no entries for BLM for 2013.
Total authorized AUMs for BLM and USFS between 2002 and 2012 averaged 14,639,848, with a range of between 12,656,540 in 2004 and 15,819,413 in 2010. The number of total AUMs was reduced due to drought in 2003 and 2004. (GAO 2005, 14)
13 Table 3: USFS Authorized Number of Permittees and AUMs, 2002-2013
Grazing Season Total Authorized Number Total of Permittees Authorized AUMs 2002 6,830 6,402,125 2003 5,638 5,725,785 2004 5,791 5,288,091 2005 6,457 6,569,171 2006 5,305 5,675,098 2007 5,344 5,920,850 2008 5,931 6,621,931 2009 6,141 6,673,526 2010 6,206 7,056,298 2011 6,014 6,799,016 2012 5,906 6,905,657 2013 5,711 6,388,964 Average 5,940 6,335,542 Sources: USDA Grazing Statistical Summaries FY 2002 to 2013 (categories of “paid permits” or “commercial livestock”).
Table 4: USFS and BLM – Permits, Leases and Permittees by Grazing Seasons, 2002-2013
Number of Permittees, Permits and Leases AUMs Grazing National BLM National BLM BLM & NFS Season Forest Authorized Forest System Authorized Total System Permits and Authorized AUMs Authorized Authorized Leases AUMs AUMs Number of Permittees 2002 6,830 15,072 6,402,125 7,670,129 14,072,254 2003 5,638 14,880 5,725,785 7,253,613 12,979,398 2004 5,791 14,867 5,288,091 7,368,449 12,656,540 2005 6,457 15,998 6,569,171 8,518,458 15,087,629 2006 5,305 15,943 5,675,098 8,558,443 14,233,541 2007 5,344 15,935 5,920,850 8,476,842 14,397,692 2008 5,931 15,935 6,621,931 8,590,864 15,212,795 2009 6,141 17,654 6,673,526 8,608,534 15,282,060 2010 6,206 16,070 7,056,298 8,763,115 15,819,413 2011 6,014 16,117 6,799,016 8,997,890 15,796,906 2012 5,906 16,044 6,905,657 8,594,442 15,500,099 2013 5,711 n/a 6,388,964 n/a n/a Average 5,940 15,865 6,335,543 8,309,162 14,639,848 Sources: USDA Grazing Statistical Summaries, FY 2002 to 2013 (category of “paid permits” or “commercial livestock”); BLM Public Land Statistics from 2004 to 2013, Tables 3.10a and b.
14 Other Federal Agencies’ AUMs
BLM and USFS are indeed the significant players in the federal grazing arena. For the purpose of comparison, all other federal agencies approved 794,000 AUMs in 2004 (GAO 2005, 17).
3. Cost of the Federal Livestock Grazing Program
a. Grazing Receipts and their Distribution
USFS and BLM combined inflation-adjusted receipts from grazing fees have declined between 2002 and 2014. As Table 5 below shows, this decline was greater for the BLM than for the USFS. Combined BLM and USFS receipts were $27.6 million in 2002; they dipped in 2004 because of the drought, rose again slightly above the 2002 level in 2006, and from there decreased to $18.5 million in 2014.
Not all of the grazing fees that are collected get returned to the U.S. Treasury. Federal law requires that 50 percent of all grazing revenue, for both BLM and USFS, goes into range rehabilitation and improvement funds. Activities that can be funded include, but are not limited to, constructing fences to contain livestock, installing water tanks, building impoundments to improve access to water for livestock, and seeding to improve vegetation and forage. Half of these funds are designated for use in the district, region or national forest from which they were generated. The remaining half is designated for use as directed by the secretary. (GAO 2005, 31) Counties receive between 12.5 percent and 50 percent and the balance, between 25 percent and 37.5 percent gets returned to the U.S. Treasury. (Table 6)
Table 5: USFS and BLM Grazing Receipts 2002-2014, in 2014
Receipts*
USFS Grazing BLM Grazing Receipts USFS and BLM Total Fiscal Year (2014 Dollars) (2014 Dollars) (2014 Dollars)
2002 $7,889,000 $19,733,000 $27,622,000 2003 $6,169,000 $15,985,000 $22,154,000 2004 $7,010,000 $8,191,000 $15,201,000 2005 $8,397,000 $11,787,000 $20,184,000 2006 $7,620,000 $20,321,000 $27,941,000 2007 $6,514,000 $18,006,000 $24,520,000 2008 $6,247,000 $17,538,000 $23,785,000 2009 $6,148,000 $19,284,000 $25,432,000 2010 $6,005,000 $16,189,000 $22,195,000 2011 $5,974,000 $15,624,000 $21,598,000 2012 $5,689,000 $15,352,000 $21,041,000 2014 $5,300,000 $13,217,000 $18,517,000 2013 $5,027,000 $14,585,000 $19,612,000 *The numbers are gross receipts, i.e. payments to counties have not been subtracted. Sources: USDA Forest Service, Budget Overviews for Fiscal Years 2004, 2006, 2007, 2008, 2010-2015, Budget Justification Fiscal Years 2005, 2009 (2012 and 2013 numbers are “Enacted”; 2014 numbers are “Estimated.”); BLM Fiscal Years 2004 to 2015 Bureau Highlights - Appendices (Receipts by Source Category). BLM grazing receipts for 2014 “Estimated.” All other receipts are “Actual.”
15 Table 6: Distribution of Fee Receipts by Agency and Land Classification Payments to Range US Counties Betterment/Improvement Fund Treasury
USFS 25% 50% 25%
BLM Section 3 12.50% 50% 37.50% (permits)
BLM Section 15 50% 50% (leases) BLM Bankhead- 25% 50% 25% Jones (grasslands) Sources: GAO 2005, 31-32; BLM 2012.
b. Grazing Appropriations
Direct grazing appropriations, funds designated for use for BLM or USFS rangeland/grazing management programs and range improvement/betterment funds, are shown in Table 7. USFS inflation-adjusted appropriations have increased, whereas BLM appropriations have decreased. BLM and USFS combined inflation-adjusted appropriations have decreased somewhat since 2003, with the exception of the year 2012, when they increased to almost their 2004 level.
Table 7: USFS and BLM Direct Grazing Appropriations 2002-2014, in 2014 Dollars
Appropriations Fiscal Year USFS BLM USFS and BLM (2014 Dollars) (2014 Dollars) (2014 Dollars)
2002 $50,040,000 $105,948,000 $155,988,000
2003 $55,998,000 $104,771,000 $160,768,000 2004 $59,012,000 $101,369,000 $160,381,000 2005 $59,950,000 $93,989,000 $153,940,000 2006 $58,831,000 $92,214,000 $151,045,000
2007 $56,877,000 $87,794,000 $144,671,000
2008 $55,480,000 $90,830,000 $146,310,000 2009 $57,126,000 $88,213,000 $145,340,000 2010 $56,190,000 $88,751,000 $144,941,000 2011 $53,751,000 $89,343,000 $143,094,000
2012 $59,248,000 $99,616,000 $158,864,000
2013 $54,245,000 $86,495,000 $140,740,000 2014 $58,356,000 $85,280,000 $143,636,000
Sources: USDA Budget Overview 2004, 2006-2008, 2010-2015; USDA Budget Justification 2005, 2009; BLM Bureau Highlights 2004-2015. NOTE: USFS numbers were reported as “Enacted” for the years 2004 and 2007- 2014, and for other years as “Final Appropriations.” BLM 2012 and 2014 numbers were reported as “Enacted”; in 2013, the number for range improvement, which is part of the total, was reported as “Requested.” For other years, BLM numbers were reported as “Actual.”
16 c. Difference Between Appropriations and Receipts In FY 2014, the total inflation-adjusted appropriations for BLM and USFS were $143.6 million; grazing receipts amounted to $18.5 million, or 13 percent of the appropriations (Table 8 and Figure 4).
The difference between appropriations and receipts was $128.4 million in 2002, and reached its highest level of $145.2 million in 2004. Its lowest leve1of $120.2 million was reached in 2007. The percent of receipts to appropriations was 18 percent in 2002, dipped to 9 percent in 2004, rose again to18 percent in 2006, and decreased to 13 percent in 2013.
The difference between appropriations and receipts is a measure of the cost to taxpayers of the grazing program. The direct federal subsidy of the BLM and USFS livestock grazing programs exceeded $120 million every year for the past 12 years.3
Table 8: Total BLM and USFS Direct Grazing Appropriations vs. Receipts 2002-2014
Year Total Total Receipts Appropriations Percent Receipts to Appropriations (2014 Dollars) Minus Receipts Appropriations (2014 Dollars) (2014 Dollars) Percent of Receipts to Appropriations 2002 $155,988,000 $27,622,000 $128,365,000 18% 2003 $160,768,000 $22,154,000 $138,614,000 14% 2004 $160,381,000 $15,201,000 $145,181,000 9% 2005 $153,940,000 $20,184,000 $133,756,000 13% 2006 $151,045,000 $27,941,000 $123,104,000 18% 2007 $144,671,000 $24,520,000 $120,151,000 17% 2008 $146,310,000 $23,785,000 $122,525,000 16% 2009 $145,340,000 $25,432,000 $119,907,000 17% 2010 $144,941,000 $22,195,000 $122,746,000 15% 2011 $143,094,000 $21,598,000 $121,496,000 15% 2012 158,864,000 $21,041,000 $137,824,000 13% 2013 140,740,000 $19,612,000 $121,127,000 14% 2014 143,636,000 $18,517,000 $125,119,000 13%
3 Federal receipts are gross numbers, before distribution to counties, for this comparison.
17 Figure 4
As shown, inflation adjusted grazing appropriations for USFS have increased, whereas they have decreased for BLM (Table 7). However when comparing grazing receipts as a percent of appropriations separately for BLM and USFS (Tables B3 and B4), different trends emerge.
The percentage of receipts to appropriations declined markedly for the Forest Service, from 16 percent in 2002 to 9 percent in 2014. (Table B3)
The changes in the percentage of receipts to appropriations are less pronounced for BLM, because appropriations have fallen as well as receipts. The percentage of receipts to appropriations for BLM was 19 percent in 2002 and is down to 15 percent in 2014, but fluctuated a lot in-between and rose to 22 percent in both 2006 and 2009. (Table B4) d. PRIA Fee Impacts on Receipts and Scenarios for Reducing Taxpayer Costs
As demonstrated above, BLM and USFS grazing receipts have declined in real, or inflation adjusted dollars, since 2002. The decline in grazing receipts over these years is mostly attributable to the decline in grazing fees rather than a decline in the number of AUMs. The grazing fee is based on a specific formula called PRIA and explained in the next chapter.
Specifically, the inflation-adjusted PRIA fee was $1.88 in 2002, and $1.36 in 2013 (Table B6), a decline of 28 percent. Grazing receipts in 2002 were $27.6 million and in 2013 they were $19.6 million (Table 7), a decline of 29 percent. Receipts decline when either the numbers of AUMs
18 fall in any given year, or fees decline. As Tables 2 and 3 show, AUMs have been fairly stable over the years, except for the drought period around 2004.
Before the PRIA formula came into effect, BLM and USFS charged fees that were designed to cover agency costs or were market-based.
In order to cover direct appropriation costs for the BLM and USFS programs, the grazing fee would have had to be set at $10.25 per AUM based on 2012 figures. This calculation is based on using the 2012 inflation-adjusted appropriations for BLM and USFS of $158.9 million (Table 7), and total AUMs of 15.5 million (Table 4). 4
Other federal agencies as well as state agencies administering grazing programs in the West sometimes charge rates that are considerably higher than the $10.25 per AUM.
If the BLM and USFS had charged private markets rates, on average between 2002 and 2012, grazing receipts would have amounted to $261 million annually. This is based on applying private grazing fee rates to non-irrigated land to the corresponding number of AUMs for each of those years. (Table B5)
Specifically, applying the inflation-adjusted 2012 private grazing fee of $19.23 per AUM, grazing fees would have been $298 million, way above the appropriations of $158.9 million for that year. (Table B5)
4. BLM and USFS Grazing Fees a. History of Fee Setting Approaches
Charging fees for grazing livestock has been Forest Service policy since 1906. The BLM and its predecessors have charged fees since 1936, after enactment of the Taylor Grazing Act in 1934. (GAO 2005, 84)
The USFS and BLM have used a number of different concepts at different points in time to establish how much the federal government should charge livestock operators for grazing their animals on federal lands:
• Fair market value: The fee is determined by the interaction of willing buyers and sellers that are knowledgeable about the value of what they are transacting, behave in their own best interest, and are free of undue pressure to conduct the transaction. A knowledgeable buyer and seller would be aware of the quality and productivity of the land and of market prices for leases or permits on other comparable lands, and the seller would expect to derive some profit from the transaction.
• Costs to the government: The fee is determined with the goal of covering the costs to the government of providing land for grazing.
4 2012 is the most recent year where combined USFS and BLM AUMs data are available (Table 6).
19
• Fees based on the development of livestock prices: Fees increase or decrease depending on livestock prices.
• PRIA Fees: Fees are set to achieve a policy objective of supporting the livestock industry (“prevent economic disruption to the Western livestock industry” (43 U.S.C. 37, §1901(5)), taking into account the price for beef and lamb and the costs of production for the livestock industry. Costs of the public agency for providing grazing land are not included in this fee structure.
In the early 1900s, the Forest Service assessed fees by comparing those of similar privately owned rangeland, so as to approximate fair market value. From the early 1920s to 1968, the USFS based its fees on beef and lamb prices. BLM and its predecessors started out basing their fees on agency expenses, but changed this approach in 1958 to basing fees on livestock prices, similar to the USFS fee structure. (GAO 2005, 84) In the 1960s, the Bureau of the Budget (predecessor of Office of Management and Budget) set a fee schedule for these two agencies that had the goal of achieving fair market value. (GAO 2005, 84) This was based on an Office of Management and Budget circular of 1959, which directed that “fair market value” be obtained (36 C.F.R. §222.50 (b)).
36 CFR §222.50 General procedures.
(b) Guiding establishment of fees are the law and general governmental policy as established by Bureau of the Budget Circular A-25 of September 23, 1959, which directs that: A fair market value be obtained for all services and resources provided the public through establishment of a system of reasonable fee charges, and that the users be afforded equitable treatment. This policy precludes a monetary consideration in the fee structure for any permit value that may be capitalized into the permit holder's private ranching operation.
The Public Rangeland Improvement Act (PRIA) of 1978 acknowledged that the public rangelands were in unsatisfactory condition with regard to productive potential for livestock, wildlife and wildlife habitat, recreation, forage and soil conservation, and might decline further under prevailing management. It acknowledged the need for intensive programs for maintaining, managing and improving these lands. (43 U.S.C. 37, §1901(1)-(4) In further elaborating the policy objectives, it stated that “to prevent economic disruption and harm to the Western livestock industry, it is in the public interest to charge a fee for livestock grazing permits and leases on the public lands which is based on a formula reflecting annual changes in the costs of production.”(43 U.S.C. 37, §1901(5))
The Public Rangeland Improvements Act (PRIA) established a fee formula on an experimental basis (to be applied for the grazing years 1979 through 1985), explained in 43 U.S.C. 37 §1905:
For the grazing years 1979 through 1985, the Secretaries of Agriculture and Interior shall charge the fee for domestic livestock grazing on the public rangelands which Congress finds represents the economic value of the use of
20 the land to the user, and under which Congress finds fair market value for public grazing equals the $1.23 base established by the 1966 Western Livestock Grazing Survey multiplied by the result of the Forage Value Index (computed annually from data supplied by the Economic Research Service) added to the Combined Index (Beef Cattle Price Index minus the Price Paid Index) and divided by 100: Provided, That the annual increase or decrease in such fee for any given year shall be limited to not more than plus or minus 25 per centum of the previous year’s fee.
In 1981, USFS and BLM began charging the same fees based on this fee formula. After the seven-year trial period, President Reagan issued Executive Order 12548 (Feb. 14, 1986) to continue the PRIA fee formula indefinitely, and established a minimum fee of $1.35 per AUM. (Vincent 2012, 3)
The PRIA fee formula is applied only for grazing on western USFS forestlands and permits and leases on BLM lands. Detailed regulations for western states can be found in 36 CFR 222.51, and in 43 CFR 4130.8-1.
Fees for grazing on USFS Grasslands are regulated by 36 CFR 222.52, which states that: “Grazing fees for National Grasslands will be established under concepts and principles similar to those in §222.51.” These fees are usually slightly higher than the fees based on the PRIA formula. (GAO 2005, 39)
Grazing on USFS lands in non-western states is regulated either by 36 CFR 222.53, setting out non-competitive procedures based on fair market value, or they are regulated by 36 CFR 222.54, which provides for determining fees by competitive procedures. Other federal agencies base their grazing fees mostly on market value. (GAO 2005, 41-44)
Table 9: PRIA-Based Grazing Fees from 1981 to 2014
Dollars per AUM 1981...... $2.31 1993...... $1.86 2005...... $1.79 1982...... $1.86 1994...... $1.98 2006...... $1.56 1983...... $1.40 1995...... $1.61 2007...... $1.35 1984...... $1.37 1996...... $1.35 2008...... $1.35 1985...... $1.35 1997...... $1.35 2009...... $1.35 1986...... $1.35 1998...... $1.35 2010...... $1.35 1987...... $1.35 1999...... $1.35 2011...... $1.35 1988...... $1.54 2000...... $1.35 2012...... $1.35 1989...... $1.86 2001...... $1.35 2013...... $1.35 1990...... $1.81 2002...... $1.43 2014...... $1.35 1991...... $1.97 2003...... $1.35 1992...... $1.92 2004...... $1.43
Sources: Vincent 2012, 3; BLM 1/31//14; BLM 2013
21 Actual grazing fees charged since 1981 based on the PRIA formula for USFS and BLM permits and leases are shown in the table below. Several attempts at legislative reform have been made since 1986 with the goal of changing the PRIA formula to bring grazing fees charged on federal lands closer to rates charged for private and state grazing lands in the western states. None of these attempts were successful. (Vincent 2012, 5-6)
For the years 2002 to 2014, Table 9 shows that the PRIA fee was $1.43 in 2002, rose to its highest level of $1.79 in 2005, and then fell to the its $1.35 legal minimum level for the years 2007 to 2014.
When adjusted for inflation, PRIA fees have declined from $1.88 in 2002 to $1.35 in 2014, with a high of $2.20 in 2005 (in 2014 dollars). (Table B6)
b. PRIA Formula Explained
A 2012 report by Vincent explains that the PRIA formula is to represent the fair market value of grazing, beginning with a 1966 base value of $1.23 per AUM. This value is adjusted for three factors: (1) the rental charge for pasturing cattle on private rangelands (FVI), (2) the sales price of beef cattle (BCPI), and (3) the cost of livestock production (PPI). (Vincent 2012, 3) The values for FVI, BCPI and PPI that were used to calculate PRIA fees from 1981 to 2014 are shown in Table 12 below.
The following formula is used to determine the grazing fee for any given grazing season:
The example below describes the application of the PRIA formula.
Applying index numbers for data year 2013 (Table 12) to calculate the PRIA fee for 2014 would result in a calculated fee (CF) of $0.75.
2014 CF = $1.23 x (507+548-994)/100 = $0.75
Since $1.35 has been set as a minimum value, the actual grazing fee was $1.35 for 2014.
CF = Calculated Fee is calculated fee (CF) is the value resulting from the application of the above formula and is based on prior year values for the formula components (FVI, BCPI, PPI). The actual fee may differ from the calculated fee. If the calculated fee is below the minimum of $1.35 per AUM, then the actual fee will be set at $1.35. Or, if the fee adjustment would exceed 25 percent of the previous year’s fee, then the fee would be constrained within the 25 percent limit.
$1.23 was the base year fee for forage for the year 1966. The base year fee of $1.23 was calculated as the difference between the costs of conducting ranching business on private lands, including any grazing rates charged, and the costs of ranching on public lands, not including
22 grazing fees. The costs were computed in a 1966 study that included 10,000 ranching businesses in the western states. (GAO 2005, 40, FN 27)
FVI = Forage Value Index is based on grazing rates charged per head month on privately owned, non-irrigated land in eleven western states.5 Private grazing rates – for the prior year – are published by USDA’s National Agricultural Statistics Service (NASS) in the January Agricultural Prices report. They are divided by the base year rate ($ 3.65) times 100, to arrive at the Index Value.
BCPI = Beef Cattle Price Index is based on weighted average annual selling price for beef cattle in 11 western states of the prior year. It is published by USDA’s NASS in the December Agricultural Prices report. The beef cattle price is divided by the base year price ($22.04) times 100 to arrive at the index.
PPI = Prices Paid Index is based on several categories of livestock production costs and is published by USDA’s NASS in the December Agricultural Prices report. (43 CFR 4130 .8–1)
As explained above, based on a 1966 study, it was determined that a public lands grazing fee of $1.23 would make total grazing costs on public land (non-fee costs + fee of 1.23) equal to total costs on private land (including lease rates for private grazing land). This $1.23 fee was at the time 33.69 percent of the private grazing fee, which amounted to $3.65.
The PRIA formula has been used to adjust federal grazing fees yearly, and has resulted in an ever-widening gap between private rates and PRIA fees, as will be shown in more detail below. PRIA fees have declined since 1981 in both nominal terms and adjusted for inflation, whereas private fees have increased nominally and changed little in real terms. (Table B6)
This development is the result of the construction of the PRIA formula. In the nominator it adds the difference between the BCPI (Beef Cattle Price Index) and PPI (Prices Paid Index) to FVI (Forage Value Index). PPI has mostly been larger than the BCPI, as can be seen in Table 10. This results in a negative BCPI-PPI number, which reduces the nominator below the level of FVI, and therefore reduces the factor by which the base fee is multiplied. With FVI thus being systematically reduced, it is not possible for the PRIA fee to increase in sync with private rates. The calculated fee can even drop below the 1.23 base fee or become negative (Table 10). Only the legally established minimum floor of $1.35 prevents that.
If the fee formula had been CF = 1.23 X FVI/100, public grazing fees would have increased about five-fold since 1966, from $1.23 to about $6.15. This can be seen in Table 10, where FVI starts at a level of 100 in 1966 and reaches a level of 507 in 2013, resulting in a five-fold increase in private forage rates. The public grazing fee would still be about 1/3 of the private grazing fee, just as it was in 1966, when the base fee of $1.23 was established.
5 Montana, Idaho, Wyoming, Colorado, New Mexico, Arizona, Utah, Nevada, Washington, Oregon, and California
23 5. PRIA Fees Compared To Private, State and Other Federal Grazing Fees
Several grazing fee studies and surveys have been conducted over the years, demonstrating that rates for private grazing lands in the western states are generally higher, and sometimes considerably higher, than fees based on the PRIA formula. Fees charged for state grazing lands also tend to be higher than PRIA rates, though they generally are lower than private rates. Livestock operators also generally pay rates that are higher than PRIA rates on federal lands administered by other federal agencies, including National Park Service or U.S. Fish and Wildlife Service.6
Tables 10 to 13 show how PRIA fees are different from rates charged by private, state and other federal landowners. One of the differences, besides PRIA being generally lower, is that PRIA fees are uniformly applied across the western states, while it is obvious when looking at non- PRIA rates (i.e. rates charged by private, state and other federal land-owners) that they can vary considerably within and across states. This is to be expected if supply and demand conditions are not uniform, and market-based fee setting approaches are used. Supply conditions include such factors as quality of the forage, access to water, grazing infrastructure, and services offered (or not offered) by private or public landowners. Differences in supply conditions can lead to differences in livestock management costs, and in turn affect how much a rancher is willing to pay for a lease or permit. Livestock operator costs are also not uniform on BLM and USFS land. A 1992 study for the states of Idaho, New Mexico and Wyoming showed that total non-fee costs for livestock operators (excluding the federal grazing fee) on BLM land on average were $15.41 per AUM and on USFS land were $21.89 per AUM. (VanTassel, 2012) a. PRIA Fees Compared to Private Fees
Table 10 includes a column shaded in light grey, which shows the average private grazing fees (for non-irrigated land in the 16 western states) from 1981 to 2013. The column on the right calculates PRIA fees as a percent of private land grazing fees. Going back to the PRIA base year 1966, the calculated fee would have been 33.69 percent of the private fee.
Starting in 1981, when PRIA fees were first charged by both BLM and USFS, the PRIA fee was down to 23.79 percent of the private fee, the highest it has been since then. By 1984 the percentage had gone down to 14.33 percent, but it rose sporadically for a few years after that. Since 1991 it has trended downward, reaching a low of 6.72 percent in 2013.
6 Van Tassell et al. 1997; Torell et al. 1993; Torell et al. 2003; Van Tassell 2012; Bioeconomics 2011, GAO 2005.
24 Table 10: PRIA Fee Calculation 1980 through 2014 and Comparison to Private Rates
PRIA Data Calculated PRIA Private % PRIA Fee of FVI BCPI PPI Fee Year Fee (CF) Fee Rates Private Rates Year
1966 100 100 100 $1.23 $3.65 33.69%
1980 216 291 319 1981 $2.31 $2.31 $9.71 23.79% 1981 242 268 359 1982 $1.89 $1.86 $9.75 19.08% 1982 229 262 378 1983 $1.39 $1.39 $9.59 14.49% 1983 242 256 387 1984 $1.37 $1.37 $9.56 14.33% 1984 243 262 395 1985 $1.35 $1.35 $9.06 14.90% 1985 251 243 397 1986 $0.93 $1.35 $8.33 16.21% 1986 233 235 388 1987 $0.98 $1.35 $8.09 16.69% 1987 234 272 381 1988 $1.54 $1.54 $8.98 17.15% 1988 240 297 386 1989 $1.86 $1.86 $10.06 18.49% 1989 243 306 402 1990 $1.81 $1.81 $10.86 16.67% 1990 253 326 419 1991 $1.97 $1.97 $9.78 20.14% 1991 265 327 436 1992 $1.92 $1.92 $10.46 18.36% 1992 275 316 440 1993 $1.86 $1.86 $10.60 17.55% 1993 279 333 451 1994 $1.98 $1.98 $11.30 17.52% 1994 282 304 455 1995 $1.61 $1.61 $11.20 14.38% 1995 301 277 473 1996 $1.29 $1.35 $11.40 11.84% 1996 293 252 499 1997 $0.57 $1.35 $11.70 11.54% 1997 310 281 512 1998 $0.97 $1.35 $12.30 10.98% 1998 323 272 514 1999 $1.00 $1.35 $12.30 10.98% 1999 326 281 516 2000 $1.12 $1.35 $12.60 10.71% 2000 329 313 554 2001 $1.08 $1.35 $13.10 10.31% 2001 345 330 559 2002 $1.43 $1.43 $13.50 10.59% 2002 356 303 559 2003 $1.23 $1.35 $13.80 9.78% 2003 367 342 593 2004 $1.43 $1.43 $14.30 10.00% 2004 378 402 618 2005 $1.99 $1.79 $14.60 12.26% 2005 400 413 686 2006 $1.56 $1.56 $15.20 10.26% 2006 414 418 724 2007 $1.32 $1.35 $15.90 8.49% 2007 427 394 762 2008 $0.73 $1.35 $16.20 8.33% 2008 444 394 891 2009 $-0.65 $1.35 $16.10 8.39% 2009 433 355 806 2010 $-0.22 $1.35 $16.70 8.08% 2010 441 398 866 2011 $-0.33 $1.35 $17.90 7.54% 2011 460 509 946 2012 $0.28 $1.35 $18.80 7.18% 2012 490 556 980 2013 $0.81 $1.35 $20.10 6.72% 2013 507 548 994 2014 $0.75 $1.35 n/a* n/a Source: USDA National Agricultural Statistics Service (NASS) Quickstats; USDA NASS Agricultural Prices (January and December Surveys); GAO 2005, 83; correspondence with BLM rangeland specialist; USDA NASS 2007.
25 Figure 5
Figure 6
26 Table 11 compares PRIA fees to average private grazing fees per AUM in 16 western states for the years 2002 and 2013. The state averages for private grazing fees varied considerably among the 16 states. In 2002, rates ranged from $7.30 to $20.9. In 2013, fees varied between $9 and $33.50 per AUM.
How much private rates can vary even within one state is exemplified by a 2013 Colorado State University Extension survey. Rates for privately owned, non-irrigated pasture leases for Colorado averaged $16.49 per AUM, and ranged from $3.72 and $38 per AUM. 7
Table 11: Private Rates and PRIA Fees per AUM for Non-Irrigated Land In 16 Western States, 2002 and 2013 PRIA Fee 2002 PRIA Fee 2013
$1.35 $1.35 State Average Private Average Private Fee
Fee 2002 2013
Arizona $7.30 $9.00 California $12.80 $19.50
Colorado $12.60 $17.50 Idaho $11.70 $15.50 Kansas $13.00 $17.00
Montana $15.10 $21.00 Nebraska $20.90 $33.50 Nevada $10.50 N/A
New Mexico $8.80 $13.00
North Dakota $12.50 $18.00 Oklahoma $7.50 N/A Oregon $11.80 $15.00
South Dakota $16.90 $27.90 Utah $11.60 $14.50 Washington $9.60 $13.50
Wyoming $13.50 $18.70
Sources: Private rates provided by National Agricultural Statistics Quickstats
b. Fees Charged by other Federal Agencies
In 2005, the Government Accountability Office (GAO) undertook an extensive survey of grazing rates. Included in the survey were the PRIA based rates charged by BLM and USFS and the rates charged in western states by other federal agencies that are not subject to the PRIA formula. The lowest rate charged by other federal agencies in 2004 was $0.29 per AUM, charged by the U.S. Fish and Wildlife Service, the highest was $112.50 charged by the U.S. Corps of Engineers. (GAO 2005, 39)
7 Tranel et al. 2013,Table 7, 9.
27 A number of federal agencies including the National Park Service, U.S. Fish and Wildlife Service, Bureau of Reclamation, and Department of Defense set fees that reflect, or come close to, market value.
Some of the agencies, such as the Air Force and National Park Service, charge per acre; and others, such as the Corps of Engineers, receive a total bid price for an allotment. To achieve a fair market value, some agencies use a competitive bidding process that can involve notifying the public of the opportunity to permit or lease a grazing allotment, the acceptance of sealed bids, and the selection of the highest bid. Other agencies conduct a market appraisal of a grazing property, or use an average prevailing rate for the local area, and set a fee based on those values. (GAO 2005, 41-42) USFS applies market-based methods for determining fees in the eastern national forest lands. (36 CFR 222.53, 36 CFR 222.54)
Table 12: Fees Charged by Other Federal Agencies in 2004 Agency Range of fees charged Average Fee Charged Per AUM DOI National Park Service $1.35 to $7.00 $4.30 ($1.50 to $25.00 per acre) Bureau of Reclamation $1.27 to $56.46 $10.93 U.S. Fish and Wildlife Service $0.29 to $34.44 $11.24
DOE $1.43 DOD Air Force $1.35 to $26.67 $15.49 Army $0.99 to $66.09 $19.10 Corps $0.82 to $112.50 $6.22 Navy $10.42 to $97.49 $32.60 PRIA Fee 2004 $1.43 Source: GAO 2005, 39.
c. Fees Charged by State Agencies
There is also considerable variation in fees charged and fee setting approaches used by state agencies that manage public grazing lands in 17 western states.8 Table 13 shows averages and ranges of fees charged for the years 2004 and 2010. As the 2005 GAO study reports, six states (Montana, Nebraska, New Mexico, North Dakota, Oklahoma, and South Dakota) offer their leases to the highest bidder through a competitive process. Six states (Arizona, California, Colorado, Texas, Washington and Wyoming) use market-based approaches. These rely on regional market rates, land appraisals, or formulas that adjust the market price and account for differences between state and private lands and livestock market conditions. Three states (Idaho, Oregon and Utah) use formulas that apply either a base fee, adjusted for local livestock market conditions, or a fixed percentage of livestock production value.” (GAO 2005, 93-94)
8 GAO included 17 western states, including the 16 states in which the PRIA fees apply, plus Texas.
28 Table 13: Grazing Fees Charged by State Land Agencies in Western States in 2004 and 2010
State 2004 State Land Agencies Fees 2010 State Land Agencies Fees
Arizona $2.23 $2.28 California $1.35 to $12.50 No set fee
Colorado $6.65 to $8.91 35% less than private Idaho $5.15 $5.12 Kansas * * Montana $5.48 to $80.00 Minimum of $6.12 Nebraska $16.00 to $28.00 $22.50 to $38.00 Nevada * * New Mexico $0.71 to $10.15 per acre $3.19 North Dakota $1.73 to $19.69 per acre Set by auction Oklahoma $7.00 to $16.00 $8.34 to $20.83 Oregon $4.32 $5.30 South Dakota $3.00 to $56.00 per acre $10.82 Texas $4.16 to $12.50 $65.00 to $150.00 Utah $1.43 or $2.35 3.92 to $7.00 Washington $5.41 or $7.76 $8.78 Wyoming $4.13 $4.64 PRIA Rate $1.43 $1.35 * Kansas and Nevada do not have grazing on state trust lands. Sources: GAO 2005, 45-46 (2004 fees); Bioeconomics 2011, 9 (2010 fees)
As part of the 2005 GAO study, state officials were interviewed and maintained that their state trust lands, received from the federal government to generate revenue for public education, are often comparable in range condition, productivity and land value to federal lands. For example, in Wyoming and Oklahoma, state lands are intermingled with or adjacent to federal lands and have similar characteristics. (GAO 2005, 93)
6. Indirect Costs Of Federal Public Lands Grazing
The direct budgeted BLM and USFS costs as shown in the body of this paper are not the only costs of the federal grazing program. They are however the clearly measurable ones.
The USFS and BLM administer many other programs and projects not budgeted under grazing or range management budget line items that nevertheless may benefit public lands livestock operators or compensate for the negative impacts of livestock grazing. The funds spent on these activities are indirect costs of the BLM and USFS grazing programs. An example of a BLM program that benefits livestock operators is the Wild Horse and Burro Management Program that removes competitors from the range; USFS has a vegetation and
29 watershed management program that deals, for example, with soil erosion, stream degradation and weed infestations resulting from livestock grazing. For more examples of USFS and BLM programs supporting livestock grazing or compensating for its impacts see Tables A1 and A2.
In addition, there are other federal agencies that have programs benefitting grazing or attempting to compensate for its impacts. The U.S. Fish and Wildlife Service, for example, conducts consultations with USFS and BLM regarding impacts of grazing on threatened or endangered species, and for recovery plans developed for such species. For more examples see Table A3. Examples for how potential livestock grazing impacts on the state and local level see Table A4.
Indirect costs may be divided into tangible and intangible. Tangible costs are easily expressed in monetary terms, such as federal funding for fire suppression necessitated by grazing impacts. (Table A2)
Intangible costs, such as the loss of an endangered species due to public lands grazing, damage to ecosystems, or the destruction of archeological, historical and cultural resources of the indigenous people of the West, are not readily expressed in monetary terms. 9 The costs of such damage or are often “externalized,” that is, paid by communities or individuals that would otherwise benefit from the goods and services that flow from unimpaired forests, grasslands, deserts and rivers, or from the preservation of their heritage.
In this report, we do not attempt a comprehensive review of the subject or provide an estimate of the dollar value of indirect costs of the federal livestock program. A comprehensive analysis of the indirect costs of the federal grazing program is long overdue.
Conclusions
Since 2002, on average, 14.6 million AUMs have been authorized annually for grazing on USFS and BLM lands in the West. Inflation-adjusted federal appropriations for the management of the livestock grazing programs have decreased over this time period. Grazing receipts have also decreased—but at a steeper rate than the decline in appropriations. This is reflected in the percentage of receipts to appropriations, which has decreased from 18 percent in 2002 to 13 percent in 2014.
Grazing receipts have decreased mainly because the PRIA fee has decreased by 28 percent between 2002 and 2014 in inflation-adjusted dollars.
9 Economists have developed a variety of methods for estimating the monetary value of or damage to ecosystems, as well as other intangible costs. See the website http://www.ecosystemvaluation.org/dollar_based.htm for an overview of such approaches. One method is to use monetary costs resulting from damage to ecosystems as a proxy for estimating externalized, intangible, ecological and social costs. For example, the costs of eradicating invasive species and recreating conditions favorable for native re-vegetation can be used as an estimate of the value of these native plant communities.
30 There is also a growing disparity between BLM and USFS grazing fees and the rates charged for grazing on private non-irrigated land in the West. The percentage of federal fees to private rates has decreased to 6.72 percent in 2013. This decline is the continuation of a long-term decline that started right after PRIA fees were first applied by both BLM and USFS in 1981. The PRIA fee then was 23.79 percent of the private rate.
The result of this growing gap between public land fees and private rates is that livestock operators on BLM and USFS lands pay significantly less than operators on non-irrigated private rangeland, indicating a deepening of federal support for the livestock operators who have permits and leases for grazing on USFS and BLM lands.
Additionally, the PRIA fee is uniformly applied to BLM (11 western states) and USFS (16 western states) grazing leases and permits. This is a markedly different approach from fees charged by other federal agencies, and fees charged on state trust lands and private lands. Many federal agencies and state agencies use market-based fee-setting approaches that lead to considerable variations in fees within and between states, besides being generally higher than PRIA fees. Given that grazing fees are offsetting smaller and smaller portions of the appropriations, and the disparity between private rates for non-irrigated lands and the federal fee have increased considerably, the fee-setting approach for BLM and USFS lands requires a reevaluation.
31
Appendix A
Table A1: BLM Budget Items Potentially Containing Indirect Costs of Grazing BLM
Budget Item* Explanation
Soil, Water, and Air Livestock are the principal cause of soil erosion and stream degradation. Management (Jones 2000; Belsky et al, 1999) Riparian Management Livestock are the most pervasive cause of riparian damage. Up to 80% of Western streams have been damaged by livestock. (Belsky, et al. 1990) Cultural Resources Most harm to archeological resources is from livestock and from ranch Management access roads, fences, tanks and other ground disturbing range developments. (Osborn et al. 1987; Broadhead 1999) Wild Horse and Burro Removes wild competitors for forage. ** Management Wildlife and Fisheries Considerable harm to wildlife results from the pervasive competition for Management forage and removal of cover by livestock (Fleischner 1994) Threatened and Considerable harm to threatened and endangered species results from the Endangered Species pervasive competition for forage and removal of cover by livestock Management (Fleischner 1994, Flather et al. 1994;Czech and Kraussman 1997) Resource Management Grazing is a major element of planning. It covers a larger area than all other Planning uses.**
Hazardous Materials Herbicides are the main tool used to control weeds that are spread by Management livestock operations. Many noxious weeds are spread by livestock operations. (Belsky and Gelbard 2000, Reisner, 2013) Transportation and Field offices and additions are necessitated in part by the range program. Facilities Management Frequent monitoring and inspections related to grazing permits require transportation.** Workforce and Out of a workforce of 5994 full time equivalents (FTE), 670 worked directly Organizational Support for range management. (BLM Budget Justification FY 2015, VII-25,26) That amounts to 13.4% of BLM’s workforce, not counting anyone working indirectly to support livestock grazing in the administration of other programs. *Budget Categories are from BLM Bureau Highlights 2015; ** Moskowitz and Romaniello 2002, Table A-2
32 Table A2: USFS Budget Items Potentially Containing Indirect Costs of Grazing Program
Budget Item* Explanation Land Management Every 10-15 years land management plans must be revised, they must Planning include suitability analysis for grazing, and an Environmental Impact Statement. Inventory and monitoring are required of forage use and Inventory and Monitoring range condition as part of the permitting process, as well as for Forest Plans and Biological Opinions.** Recreation, Heritage and Fencing off campgrounds and archeological sites may be necessitated Wilderness largely due to livestock. (Osborn et al. 1987; Broadhead 1999)
Wildlife and Fisheries Grazing is often the land use most in conflict with wildlife habitat needs Habitat Management and necessitates fencing. (Fleischner 1994) Vegetation & Watershed Grazing is often the land use most damaging to soils and vegetation, Management causing widespread soil erosion and infestations of grasslands with weeds, shrubs and conifers.(Belsky and Blumenthal 1995, Reisner 2013) Wildland Fire Grazing is the principle cause of the growth of highly flammable Management thickets in western ponderosa pine forests, and for invasion of rangelands by pinion, juniper and other woody shrubs. Wildland fire management includes thinning of thickest and prescribed fires to reduce fuel loads. (Belsky and Blumenthal 1995) Forest and Research stations spend some of their efforts studying the impacts of Rangeland grazing on native species and ecosystems. ** Research *Budget Categories are from USDA 2015 Budget Overview, 2015; **Explanations based on Moskowitz and Romaniello, 2002, Appendix A-1
33 Table A3: Other Federal Agencies’ Indirect Costs of Grazing on Federal Lands
Federal Agency Explanation USDA Wildlife Services Kills thousands of native carnivores each year to protect livestock in the West from predators. (Moskowitz and Romaniello 2002, 24) US Fish and Wildlife Service Expends part of its budget for listing native species impacted by grazing as threatened or endangered, for consultations with USFS and BLM over impacts of grazing on listed species, and for recovery plans for such species. (Moskowitz and Romaniello 2002, 24) USDA’s livestock assistance program Payments to ranchers in the event of livestock loss due to natural disasters, like droughts. Moskowitz and Romaniello (2002, 25) report that on average public lands ranchers qualify for this subsidy four out of 10 years. Department of Commerce National Performs the functions of the Fish and Wildlife Service Marine Fisheries Service for anadromous fish like salmon that can be impacted by livestock grazing. (Moskowitz and Romaniello 2002, 25) USDA’s Natural Resources Conservation Addresses watershed damage caused by livestock. Service (Belsky et al. 1999) Bureau of Reclamation Increased soil erosion from grazing (Jones 2000; Belsky et al, 1999) can lead to increased sedimentation of waterways and reservoirs, shortening the useful life of reservoirs and causing higher peak flows, which affect the design of dams. (Moskowitz and Romaniello 2002, 25) US Army Corps of Engineers Costs of flood control related to increased peak flows and higher erosion due to compaction of soils and removal of vegetation by livestock grazing. (Belsky et al. 1999) Environmental Protection Agency Monitoring and addressing grazing impacts on water quality due to erosion. (Moskowitz and Romaniello 2002, 25) Department of Justice Costs of defending federal land agencies in lawsuits brought by grazing industry and environmental groups. (Moskowitz and Romaniello 2002, 26) USDA - Cooperative State Research, Supports USFS, BLM in the management of their range Education and Extension Service resources. (GA0 2005, 27) USDA Office of General Counsel Provides legal advice and support to USFS in managing grazing lands and permits. (GAO 2005, 27) United States Geological Survey Conducts research on the effects of grazing on plant communities, including invasive species; on runoff and erosion, on select species or groups of species, and on ecosystem health including riparian areas. GAO 2005, 28)
34 Table A4: Indirect Costs on State and Local Level
Affected Entities Explanation Recreation-related industries Recreation and the industries associated with it are negatively affected from impacts of livestock on riparian habitats, trampling of vegetation and fecal matter on the ground, waterways impacted by erosion, impacts on cultural resources, and diminished opportunities for wildlife viewing. (Moskowitz and Romaniello 2002, 29) States and Communities States and communities are impacted by increased grazing-related fire risks. (Belsky and Blumenthal 1995; Swetnam and Baisan 1994) State and Local Governments Livestock grazing assists weed invasions. (Belsky and Gelbard 2000, Reisner 2013) Weed populations on federal lands can become source populations for invasions on other lands and thus require increased weed control expenses of state and local governments. (Moskowitz and Romaniello 2002, 31) State Historic Preservation Offices Studies have shown that livestock grazing can have a Indigenous Peoples destructive impact on archeological and historical sites. (Osborne et. al 1987; Broadhead 1999) Cultural artifacts of indigenous people (an intangible cost) are destroyed as the result of cattle trampling, soil erosion, and range development measures such as bulldozing and plowing. The (tangible) costs of consultations over impacts of public lands grazing on archeological resources are borne at least in part by states through their State Historic Preservation Offices. (Moskowitz and Romaniello 2002, 32) Water Utilities Livestock borne pathogens cause illness and increase water treatment costs in the West, because of the need of monitoring and control of these pathogens by water utilities throughout the West. (Moskowitz and Romaniello 2002, 32)
35 Appendix B
Table B1: BLM Acres Grazed by State, 2004
BLM Grazing Lands (acres)
Arizona 7,955,000 California 5,672,000 Colorado 6,593,000 Idaho 10,756,000 Montana 7,839,000 New Mexico 11,533,000 Nevada 39,331,000 Oregon/Washington 12,786,000 Utah 19,321,000 Wyoming 15,917,000 TOTAL 137,703,000 Sources: GAO 2005, 15
Table B2: USFS Acres Grazed by Region, 2004
Forest Service Regions Acres Grazed Eastern 75,000 Intermountain 24,107,000 Northern 8,268,000 Pacific Northwest 11,408,000 Pacific Southwest 12,353,000 Rocky Mountain 17,129,000 Southern 675,000 Southwestern 18,908,000 Total Acres of Grazed Land in All Regions 92,924,000 Total Acres of Grazed Land in Western Regions* 92,173,000 Percent of USFS Grazed Lands in the West 99.19% *All regions other than Eastern and Southern. Source: GAO 2005, 15
36 Table B3: USFS Grazing Receipts and Appropriations 2002-2014 in 2014 dollars*
Fiscal USFS Grazing USFS Appropriations Percentage Year Appropriations Total minus Receipts Of in 2014 dollars Receipts in 2014 dollars USFS in 2014 Receipts to dollars Appropriations 2002 50,040 7,889 42,151 16% 2003 55,998 6,169 49,829 11% 2004 59,012 7,010 52,003 12% 2005 59,950 8,397 51,553 14%
2006 58,831 7,620 51,211 13% 2007 56,877 6,514 50,363 11% 2008 55,480 6,247 49,233 11%
2009 57,126 6,148 50,978 11%
2010 56,190 6,005 50,184 11% 2011 53,751 5,974 47,777 11% 2012 59,248 5,689 53,559 10%
2013 54,245 5,027 49,217 9% 2014 58,356 5,300 53,056 9%
Table B4: BLM Grazing Receipts and Appropriations 2002-2014, in 2014 Dollars Fiscal BLM BLM BLM Percent Year Appropriations Receipts Appropriations Receipts to in 2014 dollars in 2014 minus Receipts Appropriations dollars in 2014 Dollars
2002 105,948 19,733 86,215 19% 2003 104,771 15,985 88,786 15% 2004 101,369 8,191 93,178 8% 2005 93,989 11,787 82,203 13% 2006 92,214 20,321 71,892 22% 2007 87,794 18,006 69,789 21% 2008 90,830 17,538 73,292 19% 2009 88,213 19,284 68,929 22% 2010 88,751 16,189 72,562 18% 2011 89,343 15,624 73,720 17% 2012 99,616 15,352 84,265 15% 2013 85,982 14,585 71,397 17% 2014 85,280 13,217 72,063 15%
*Table B4 is derived from Tables 5 and 7
37
Table B5: Scenario of USFS and BLM Grazing Revenues with Application of Private Grazing Rates
Year Private Grazing BLM and USFS Grazing Rates AUMs Revenue (in 2014 dollars) (grazing season) ($) 2002 $17.75 14,072,254 249,782,509 2003 $17.58 12,979,398 228,177,817 2004 $17.58 12,656,540 222,501,973 2005 $17.33 15,087,629 261,468,611 2006 $17.55 14,233,541 249,798,645 2007 $17.84 14,397,692 256,854,825 2008 $17.72 15,212,795 269,570,727 2009 17.49 15,282,060 267,283,229 2010 $17.60 15,819,413 278,421,669 2001 $18.40 15,796,906 290,663,070 2012 $19.23 15,500,099 298,066,904 Averages $17.82 14,639,848 261,144,544
38 Table B6: PRIA Fees and Private Fees 1981 to 2014 (Nominal and Adjusted for Inflation)
Year PRIA Fee PRIA FEE in 2014 Private Rates Private Fee in 2014 dollars
1966 $1.23 $7.43 $3.65 $18.39
1981 2.31 $5.43 $9.71 $22.84 1982 1.86 $4.12 $9.75 $21.61 1983 1.39 $2.99 $9.59 $20.65 1984 1.37 $2.86 $9.56 $19.93 1985 1.35 $2.72 $9.06 $18.28 1986 1.35 $2.68 $8.33 $16.55 1987 1.35 $2.67 $8.09 $15.97 1988 1.54 $2.93 $8.98 $17.11 1989 1.86 $3.44 $10.06 $18.60 1990 1.81 $3.26 $10.86 $19.55 1991 1.97 $3.39 $9.78 $16.80 1992 1.92 $3.24 $10.46 $17.66 1993 1.86 $3.03 $10.60 $17.28 1994 1.98 $3.13 $11.30 $17.89 1995 1.61 $2.46 $11.20 $17.12 1996 1.35 $2.03 $11.40 $17.14 1997 1.35 $1.99 $11.70 $17.28 1998 1.35 $1.97 $12.30 $17.94 1999 1.35 $1.92 $12.30 $17.52 2000 1.35 $1.86 $12.60 $17.37 2001 1.35 $1.83 $13.10 $17.74 2002 1.43 $1.88 $13.50 $17.75 2003 1.35 $1.72 $13.80 $17.58 2004 1.43 $1.76 $14.30 $17.58 2005 1.79 $2.12 $14.60 $17.33 2006 1.56 $1.80 $15.20 $17.55 2007 1.35 $1.51 $15.90 $17.84 2008 1.35 $1.48 $16.20 $17.72 2009 1.35 $1.47 $16.10 $17.49 2010 1.35 $1.42 $16.70 $17.60 2011 1.35 $1.39 $17.90 $18.40 2012 1.35 $1.38 $18.80 $19.23 2013 1.35 $1.36 $20.10 $20.23 2014 1.35 $1.35 n/a n/a
39
Bibliography
Belsky, J. A. and J. L. Gelbard. 2000. Livestock grazing and weed invasions in the arid West. Oregon National Desert Association, Bend, OR. 31 pp. http://www.onda.org/library/papers/index.html
Belsky, J. and D. M. Blumenthal. 1995. Effects of livestock grazing on stand dynamics, and soils in upland forests of the Interior West. Conservation Biology 11, 315-327. http://www.onda.org/library/papers/index.html
Belsky, J., A. Matzke, and S. Uselman. 1999. Survey of livestock influences on stream and riparian ecosystems in the western United States. Journal of Soil and Water Conservation 54, 419-431. http://www.onda.org/library/papers/index.html
Broadhead, W. 1999. Cattle, Control, and Conservation. Cultural Resource Management 22, 31-32. http://crm.cr.nps.gov/archive/22-9/22-09-12.pdf
Bioeconomics. 4/26/2011. Montana Trust Land – Grazing Lease Rate Evaluation Analysis. Report prepared for State of Montana Department of Natural Resources & Conservation: Trust Management Division, Missoula, Montana. http://dnrc.mt.gov/trust/agm/GrazingRateStudy/Documents/GrazingReviewByBioeconomics.pdf
BLM. BLM Public Land Statistics. FY 2002 to FY 2013. Last updated 8/1/14. http://www.blm.gov/public_land_statistics/.
BLM Budget Justification. FY 2007-2015. Accessed 8/16/2014. http://www.doi.gov/budget/index.cfm
BLM. Bureau Highlights. FY 2004 -2015. Accessed 8/16/2014. http://www.doi.gov/budget/appropriations/2015/highlights/index.cfm
BLM. Bureau Highlights Appendices – Receipts by Source Category. FY 2004-2015. Accessed 8/16/2014. http://www.doi.gov/budget/appropriations/2015/highlights/index.cfm
BLM Utah. “History of Public Land Grazing." Accessed 8/11/14. http://www.blm.gov/ut/st/en/prog/grazing/history_of_public.print.html
BLM Wyoming, Casper Field Office. “The Taylor Grazing Act.” Last updated 1/13/11. http://www.blm.gov/wy/st/en/field_offices/Casper/range/taylor.1.html
40
BLM. 2012. “Grazing Fees.” Last updated 2/1/13. http://www.blm.gov/nv/st/en/prog/grazing/grazing_fees.html.
BLM. 1/31/2013. “BLM and Forest Service Announce 2013 Grazing Fee.” http://www.blm.gov/wo/st/en/info/newsroom/2013/january/NR_01_31_2012.html
BLM. 1/31/14. “BLM and Forest Service Announce 2014 Grazing Fee.” http://www.blm.gov/wo/st/en/info/newsroom/2014/january/NR_01_31_2014.html
BLM. 3/28/14. “BLM Response to PEER Press Release: Fact Sheet on the BLM’s Management.
Czech, B. and P.R. Krausman. 1997. Distribution and causation of species endangerment in the United States. Science 277, 1116. http://www.sciencemag.org/cgi/content/full/277/5329/1116
Economist. “Subsidized cow chow.” The Economist (Mar. 7, 2002): 39.
Flather, C.H., L.A. Joyce, and C.A. Bloomgarden. 1994. Species endangerment patterns in the United States, USDA Forest Service, Ft Collins. 42 pp.
Fleischner, T.L. 1994. Ecological costs of livestock grazing in western North America. Conservation Biology 8, 629-644.
GAO. 2005. Livestock Grazing, Federal Expenditures and Receipts Vary, Depending on the Agency and the Purpose of the Fee Charged. GAO-05-869. http://www.gao.gov/new.items/d05869.pdf
Gorte, R., and C.H. Vincent, L.A. Hanson, and M.R. Rosenblum. 2/8/2012. Federal Land Ownership: Overview and Data, Congressional Research Service. http://fas.org/sgp/crs/misc/R42346.pdf.
International Society for the Protection of Mustangs and Burros. “A Bit of History, BLM and Land Management.” Accessed 8/4/2014. http://www.ispmb.org/History.html;
Moskowitz, K., and C. Romaniello. 2002. Assessing the Full Cost of the Federal Grazing Program. Prepared for the Center for Biological Diversity, Tucson, Arizona.
Olson, Eric. 1997. National Grasslands Management - A Primer. United States Department of Agriculture, Natural Resource Division of the Office of the General Counsel. http://www.fs.fed.us/grasslands/resources/documents/primer/NG_Primer.pdf.
41 Osborn, A., S. Vetter, R. Hartley, L. Walsh, and J. Brown. 1987. Impacts of domestic livestock grazing on the archeological resources of Capitol Reef National Park, Utah. Midwest Archeological Center Occasional Studies in Anthropology. 136 pp.
Reisner, Michael D.; Grace, James B.; Pyke, David A.; Doescher, Paul S. Conditions favouring Bromus tectorum dominance of endangered sagebrush steppe ecosystems. Journal of Applied Ecology, 2013 (in press)
Swetnam, T. W. and Baisan, C. H. 1994. Historical fire regime patterns in the Southwestern United States since AD 1700. Pp. 11-32 in C. D. Allen (ed.) Fire effects in Southwestern forests: proceedings of the second La Mesa Fire symposium. USDA Forest Service Rocky Mountain Forest and Range Experiment Station, General Technical Report RM-GTR-286.
Tanaka, John. 10/5/2012. “Grazing Fees on Public Rangelands.” Extension. https://www.extension.org/pages/55904/grazing-fees-on-public-rangelands#.U-RoGGP1bA.
Torell, L.A., L.W. Van Tassell, N.R. Rimbey, E.T. Bartlett, T. Bagwell, P. Burgener and J. Coen. The Value of Public Land Forage and the Implications for Grazing Fee Policy. New Mexico State Univ. Res. Bull. 767. 1993. http://www.uwagec.org/farmmgt/PUBS/B767.pdf
Torell, L.A., N.R. Rimbey, L.W. Van Tassell, J.A. Tanaka and E.T. Bartlett. 2003. “An Evaluation of the Federal Grazing Fee Formula.” Journal of Range Management. 56(2003):577-584.
Tranel, Jeffrey E., Rodney L. Sharp, John Deering, and Norman Dalsted, Colorado State University Extension. Lease Rates for Privately Owned, Non-irrigated Pasture for 2013. Accessed 8/5/2014. http://www.coopext.colostate.edu/ABM/leaserates13.pdf
USDA. Budget Justification. FY 2005, 2009, 2015. Accessed 8/11/14. http://www.fs.fed.us/about-agency/budget-performance
USDA. Budget Overview. FY 2004, 2006, 2007, 2008, 2010 to 2015. Accessed 8/11/14. http://www.fs.fed.us/about-agency/budget-performance
USDA Forest Service. Grazing Statistical Summary. FY 2002-2013. Accessed 8-5-2014. http://www.fs.fed.us/rangelands/reports/;
USDA Forest Service. History of Forest Planning. http://www.fs.usda.gov/main/planningrule/history USDA, National Agricultural Statistics Services (NASS). Agricultural Prices. January and December Surveys. Accessed 8/5/2014. http://usda.mannlib.cornell.edu/MannUsda/viewDocumentInfo.do?documentID=1002;
42 USDA, National Agricultural Statistics Service. “Grazing Fees.” 2007 Arizona Agricultural Statistics, p. 14. Accessed 8/16/2014. http://www.nass.usda.gov/Statistics_by_State/Arizona/Publications/Bulletin/07bul/pdf/pg14.pdf;
USDA, National Agricultural Statistics Services (NASS). “Quickstats.” Accessed 8/5/2014. http://quickstats.nass.usda.gov/.
USDA. 1905. The Use of the National Forest Reserves. Regulations and Instructions. Issued by the Secretary of Agriculture to take effect July 1, 1905. http://www.foresthistory.org/aspnet/publications/1905_use_book/1905_use_book.pdf.
USDOI, BLM and Office of the Solicitor. 2001. The Federal Land Policy and Management Act of 1976 as Amended. Accessed 8/4/2014. http://www.blm.gov/flpma/FLPMA.pdf.
Van Tassell, L.W. June 27, 2012. “A Primer on Federal Grazing Fees.” University of Nebraska – Lincoln Extension (Cornhusker Economics). Accessed 8/5/2014. http://agecon.unl.edu/c/document_library/get_file?uuid=130586f1-4399-4631-b14b- b35ab06bae68&groupId=2369805&.pdf.
Van Tassell, L.W., L.A. Torell, N.R. Rimbey and E.T. Bartlett. 1997. “Comparison of Forage Value on Private and Public Grazing Leases.” Journal of Range Management 50(1997): 300- 306.
Vincent, Carol Hardy. 6/19/2012. Grazing Fees: Overview and Issues. Congressional Research Service. http://fas.org/sgp/crs/misc/RS21232.pdf.
43 Biological Conservation 195 (2016) 156–168
Contents lists available at ScienceDirect
Biological Conservation
journal homepage: www.elsevier.com/locate/bioc
Why do wolves eat livestock? Factors influencing wolf diet in northern Italy
Camille Imbert a, Romolo Caniglia b,ElenaFabbrib, Pietro Milanesi c, Ettore Randi b,d, Matteo Serafini c, Elisa Torretta a, Alberto Meriggi a,⁎ a Dipartimento di Scienze della Terra e dell'Ambiente, Università di Pavia, Via A. Ferrata 1, 27100 Pavia, PV, Italy b Laboratorio di Genetica, Istituto Superiore per la Protezione e Ricerca Ambientale (ISPRA), via Cà Fornacetta 9, 40064 Ozzano Emilia, BO, Italy c Parco dell'Antola, La Torriglietta, Via N.S. Provvidenza 3, 16029 Torriglia, GE, Italy d Department 18, Section of Environmental Engineering, Aalborg University, Sohngårdsholmsvej 57, 9000 Aalborg, Denmark article info abstract
Article history: Thanks to protection by law and increasing habitat restoration, wolves (Canis lupus) are currently re-colonizing Received 24 July 2015 Europe from the surviving populations of Russia, the Balkan countries, Spain and Italy, raising the need to update Received in revised form 30 December 2015 conservation strategies. A major conservation issue is to restore connections and gene flow among fragmented Accepted 2 January 2016 populations, thus contrasting the deleterious consequences of isolation. Wolves in Italy are expanding from the Available online xxxx Apennines towards the Alps, crossing the Ligurian Mountains (northern Italy) and establishing connections with the Dinaric populations. Wolf expansion is threatened by poaching and incidental killings, mainly due to Keywords: fl Canis lupus livestock depredations and con icts with shepherds, which could limit the establishment of stable populations. Scat analysis Aiming to find out the factors affecting the use of livestock by wolves, in this study we determined the composi- Feeding ecology tion of wolf diet in Liguria. We examined 1457 scats collected from 2008 to 2013. Individual scats were geno- Prey selection typed using a non-invasive genetic procedure, and their content was determined using microscopical analyses. Wolf–human conflicts Wolves in Liguria consumed mainly wild ungulates (64.4%; in particular wild boar Sus scrofa and roe deer Capreolus capreolus) and, to a lesser extent, livestock (26.3%; in particular goats Capra hircus). We modeled the consumption of livestock using environmental features, wild ungulate community diversity, husbandry charac- teristics and wolf social organization (stable packs or dispersing individuals). Wolf diet varied according to years and seasons with an overall decrease of livestock and an increase of wild ungulate consumption, but also between packs and dispersing individuals with greater livestock consumption for the latter. The presence of stable packs, instead of dispersing wolves, the adoption of prevention measures on pastures, roe deer abundance, and the percentage of deciduous woods, reduced predation on livestock. Thus, we suggest promoting wild ungulate expansion, the use of prevention tools in pastures, and supporting wolf pack establishment, avoiding lethal control and poaching, to mitigate conflicts between wolf conservation and husbandry. Published by Elsevier B.V.
1. Introduction populations of Eastern Europe from which wolves began the re- colonization of Central Europe (Ansorge et al., 2006; Linnell et al., The wolf (Canis lupus), because of its adaptability to different envi- 2005). The Spanish wolf Canis lupus signatus (2200–2300 individuals) ronments and its ability to re-colonize territories when no persecution is slowly extending its distribution (Mech and Boitani, 2003). occurs, has in just a few decades expanded its range in Europe Wolves greatly declined in Italy, surviving in two small isolated sub- (Balciauskas, 2008; Breitenmoser, 1998; Chapron et al., 2003, 2014). populations confined to the southern and central part of the Apennines. The Russian wolf population is the largest in Europe, supporting those At their nadir in the early seventies of the last century, wolves in Italy of Baltic and North-European countries, and it is contiguous with the numbered about 100 individuals (Zimen and Boitani, 1975). Since the late eighties, wolves have shown a spontaneous rapid recovery, re- colonizing all the Apennines and reaching the western Italian and ⁎ Corresponding author. French Alps (Boitani, 2000; Breitenmoser, 1998; Fabbri et al., 2007; E-mail addresses: [email protected] (C. Imbert), [email protected] Marucco and McIntire, 2010; Valière et al., 2003). (R. Caniglia), [email protected] (E. Fabbri), [email protected] The re-colonization of the Alps would be a fundamental step for wolf (P. Milanesi), [email protected] (E. Randi), [email protected] (M. Serafini), [email protected] (E. Torretta), [email protected], [email protected] conservation in Italy and Central Europe as well (Genovesi, 2002). (A. Meriggi). Moreover, the early and ongoing wolf expansion from the eastern
http://dx.doi.org/10.1016/j.biocon.2016.01.003 0006-3207/Published by Elsevier B.V. C. Imbert et al. / Biological Conservation 195 (2016) 156–168 157
Alps will predictably increase chances to originate mixed packs and diversity, husbandry characteristics, wolf grouping and habitat occu- increase the local genetic diversity as has been already described pancy behavior (stable packs or dispersing individuals). (Fabbri et al., 2014; Randi, 2011). The sub-population of wolves inhabiting the Liguria region thus 2. Material and methods plays a crucial role in assuring the linkage between the wolves of central Italy and those of the Western Alps (Fabbri et al., 2007). If 2.1. Study area this link should break, the wolf population of the Western Alps would be isolated, perhaps failing to recolonize the remaining part This research was carried out in the Liguria region, north Italy of the Alps. (44°30′16″,8°24′10″). The study area spreads over 5343 km2 including The distribution of wolves is usually determined by the abundance a part of the Northern Apennines and of the Western Italian Alps, until of its preys, environmental characteristics, and the risk associated with the border with France. The region is divided in four provinces, Imperia, the presence of humans (Eggermann et al., 2011; Jędrzejewski et al., Savona, Genoa and La Spezia, respectively from the western to the east- 2004; Massolo and Meriggi, 1998). This last point is the key problem ern part (Fig. 1). Altitude ranges from 0 to 2153 m a.s.l.; 36% of the area of wolf conservation because wolves can have a dramatic impact on is between 0 and 400 m a.s.l., 35% between 400 and 800 m, 21% between livestock breeding, affecting human attitudes that can lead to illegal 800 and 1200 m, and 8.5% more than 1200 m a.s.l. Forests cover 63.8% killing, increasing the risk of extinction (Behdarvand et al., 2014; of the region (deciduous woods: 28.8%; conifer woods: 7.1%; mixed Kovařík et al., 2014). woods: 27.9%), pastures 6.2%, agricultural areas 17.1%, and urbanized The impact of wolves on livestock is different according to geograph- areas 3.9%. Towns and villages, as well as farmlands, are concentrated ical region. In regions with a very low abundance of wild ungulates, as in on flat terrains, close to the coasts. The climate extends from Portugal and Greece, wolves feed mainly on livestock (Migli et al., 2005; Mediterranean on the coast to sub-oceanic in the mountains. The tem- Papageorgiou et al., 1994; Vos, 2000). On the other hand, in Germany at- perature extends from −2 °C in winter to 35 °C during summer. Mean tacks on livestock are rare because shepherds equip the pastures with annual precipitation ranges from 750 to 1250 mm in the west to electric fences to protect their herds and because the wild ungulate 1350–1850 in the central and eastern part of the region. On the ridge availability is high (Ansorge et al., 2006). of the mountains and in the upper part of the valleys, snow cover can In other new-recolonizing areas such as France or North Italy, wild reach more than one meter from November to April. ungulates are the main prey of wolves, but the use of livestock is still no- The wild ungulate community includes wild boar (Sus scrofa), wide- ticeable (MEEDDAT–MAP, 2008; Meriggi et al., 2011; Milanesi et al., ly distributed with high densities (21,500 individuals shot per year in 2012). Liguria, on average from 2007 to 2012), roe deer (Capreolus capreolus), Systematic research on wolf feeding ecology has been carried out abundant in particular in the central provinces (30.9 individuals per since 1987 in the Ligurian Apennines. These studies showed an increas- km2 on average from 2009 to 2012). Fallow deer (Dama dama), intro- ing use of wild ungulates in the time but also a medium–high use of live- duced for hunting, is present in the provinces of Genoa and Savona stock species as prey (Meriggi et al., 1991, 1996, 2011; Schenone et al., (10.7 and 5.8 individuals per km2 respectively). Chamois (Rupicapra 2004). Consequently, wolf presence in Liguria, as well as in other areas rupicapra) is present only in the Maritime Alps (927 individuals counted of natural re-colonization, causes a conflict with human populations on average from 2007 to 2012), while red deer (Cervus elaphus)and that perceive predator presence as a negative element that can compro- mouflon (Ovis aries musimon) are very rare in the study area (data mise a poor rural economy. Thus, wolves suffer a high mortality mainly from Wildlife Services of Imperia, Savona, Genoa and La Spezia). due to illegal killing and accidents. This situation makes the population This high availability of wild prey promoted a natural re- vulnerable and actions aimed at a greater protection of the species are colonization of the region by wolves in the late eighties, starting from required. the provinces of Genoa and La Spezia (Meriggi et al., 1991, 1996, Usually wolf populations are structured in stable packs and lone 2011; Schenone et al., 2004). Now the wolf is present in the four wolves; packs are formed by a pair of adults, by their offspring and provinces with a minimum population of 58 individuals of which 21 other related individuals (i.e. the offspring of previous years), and some- distributed in 5 packs and 37 lone wolves, estimated by genetic analyses time by adopted individuals, whereas lone wolves are erratic individ- (see Results). uals that can temporarily establish in an area without packs. In Livestock (15,000 cows and 33,900 sheep and goats) are free- general lone wolves are young dispersing from packs but they can also grazing on pastures from April to October but the grazing period can be adults moving far from their original pack because of pack disruption be expanded or reduced depending on the weather. Pastures are often or break off for several causes (killing by humans, low prey availability partly composed of shrubs and woodlots. Only few shepherds adopt and related increasing aggressiveness, natural death of the dominant prevention methods (i.e. nocturnal recovery, guardian dogs, and electric pair) (Mech and Boitani, 2003). Packs are established in areas with fences) to deter wolf attacks. high prey availability, because only a high availability of preferred prey can dampen the aggressiveness of the pack members and avoid 2.2. Data collection pack disruption (Thurber and Peterson, 1993). Dispersing and erratic individuals use the areas without wolf packs that can be considered We divided the study area in 64 isometric cells of 10 × 10 km, as a suboptimal habitats because of the low prey availability, high human trade-off between the average territory size of the wolf in Italy (Ciucci disturbance, and possibly potential problems with local people (Fritts et al., 1997; Corsi et al., 1999; Apollonio et al., 2004; Caniglia et al., and Mech, 1981). Illegal killing can break the packs, increasing erratic 2014) and sampling feasibility. In each cell, we randomly chose an wolves and reproductive pairs that can have a greater impact in itinerary among the existing footpaths according to the Tessellation particular on livestock rearing (Wielgus and Peebles, 2014). Stratified Sampling (TSS) method that permits a better distribution The objective of the present study was to determine which factors and representativeness of the random samples than a simple random influence wolf diet, in particular, the choice of livestock as prey, which design (Barabesi and Fattorini, 2013; Barabesi and Franceschi, 2011). is the first step to find solutions for wolf conservation. With this aim, We traced a total of 64 itineraries in the study area (total length = we determined wolf diet, by analyses of scats collected in the whole 287.6 km, mean ± SD = 4.5 ± 1.59 km, min. = 2.3, max. = 10.4) Liguria region from 2008 to 2013. We highlighted the factors influenc- that were covered once a season (spring: March to May; summer: ing it, i.e. years, seasons, ungulate abundance, and social structure of June to August; autumn: September to November; winter: December wolves (packs or dispersing individuals). Then we related livestock to February), from January 2008 to August 2013 searching for wolf consumption to environmental features, wild ungulate abundance and scats and signs of wild ungulate presence (tracks, sightings, rooting, 158 C. Imbert et al. / Biological Conservation 195 (2016) 156–168
Fig. 1. Provinces of Liguria region and wolf pack territories. rubbing, wallowing, resting sites, feeding and territorial marks). We iden- and blood samples obtained from wolves found dead in the study area, 5 tified wolf scats by the size, texture, shape, and their characteristic odor. of urine, 2 hairs and 389 fresh scats containing cells of intestine epithe- All signs of presence were mapped and georeferred by a Garmin GPS. lium. Small external portions of scats and clean tissue fragments were We assessed wild ungulate abundance at transect level by comput- individually stored at −20 °C in 10 vials of 95% ethanol. Blood samples ing an Index of Kilometric Abundance (IKA) for each species (number were stored at −20 °C in 2 vials of a Tris–sodium dodecyl sulfate buffer. of found signs divided by the transect length, Meriggi et al., 1991, DNA was automatically extracted using a MULTIPROBE IIEX Robotic 1996, 2015; Milanesi et al., 2012). We estimated the abundance of Liquid Handling System (Perkin Elmer, Weiterstadt, Germany) and livestock (cattle, sheep, goats, and horses) on pastures and defined the QIAGEN QIAmp DNA stool or DNeasy tissue extraction kits (Qiagen husbandry practices by direct interviews with shepherds and by the Inc., Hilden, Germany). official data of Veterinary services of the four provinces. We identified individual genotypes for samples at 12 unlinked auto- Around each transect we defined a buffer zone corresponding to the somal canine microsatellites (short tandem repeats [STR]): 7 dinucleo- potential hunting area of wolves. We used a width of 13 km, corre- tides (CPH2, CPH4, CPH5, CPH8, CPH12, C09.250, and C20.253) and 5 sponding to the average travel distance of wolves during the night to tetranucleotides (FH2004, FH2079, FH2088, FH2096, and FH2137), go from dens or resting sites to hunting sites in Italy (Ciucci et al., 1997). selected for their high polymorphism and reliable scorability for wolves In each buffer, we measured from the Corine Land Cover III level and and dogs (Caniglia et al., 2014). the Digital Elevation Models (DEM) 12 environmental variables We determined sex of samples using a polymerase chain reaction concerning the land use, altitude, aspect, and slope using Arc GIS 9.0 (PCR)–restriction fragment length polymorphism assay of diagnostic software (Appendix A). Moreover, in each buffer we defined the hus- ZFX/ZFY gene sequences (Caniglia et al., 2012, 2013, 2014). We used a bandry variables: number of livestock heads, reared species, presence first panel of 6 STR to identify the genotypes with Hardy–Weinberg or absence of prevention methods, number of used pastures, average probability-of-identity (PID) among unrelated individuals, PID = 6 3 time past on pastures and the presence of the production “Cow-Calf 8.2 × 10 , and expected full-siblings, PIDsibs = 7.3 × 10 (Mills et al., line” consisting in cows giving birth alone, in the pasture and involving 2000; Waits et al., 2001) in the reference Italian wolves. We then used a great risk of predation by wolves on newborn calves (Dondina et al., another panel of 6 STR, also selected for their polymorphism and reli- 2014; Meriggi et al., 1991, 1996). able scorability, to increase the power of admixture and kinship analy- 9 We also collected all claimed and verified cases of predation upon ses, decreasing the PID values to PID = 7.7 × 10 and PIDsibs = livestock during the study period, recording the preyed species and 3.1 × 104 (Caniglia et al., 2014). We identified maternal haplotypes by the exact location of the events. sequencing 350 base pairs of the mitochondrial DNA (mtDNA) control region, diagnostic for the haplotype W14, which is unique to the Italian 2.3. Genetic analyses wolf population, using primers L-Pro and H350 (Randi et al., 2000; Caniglia et al., 2014). We identified paternal haplotypes by typing 4 Y- From 2007 to 2013 we collected a total of 403 presumed wolf biolog- linked microsatellites (Y-STR), MS34A, MS34B, MSY41A, and MS41B ical samples for genetic analyses. The genetic samples included 6 tissue (Sundqvist et al., 2001), characterized by distinct allele frequencies in C. Imbert et al. / Biological Conservation 195 (2016) 156–168 159 dogs and wolves (Iacolina et al., 2010). We amplified autosomal and 2.4. Diet analysis Y-linked STR loci in 7 multiplexed primer mixes using the QIAGEN Multiplex PCR Kit (Qiagen Inc.), a GeneAmp PCR System 9700 Ther- All the scats found on itineraries were preserved in PVC bags at mal Cycler (Applied Biosystems, Foster City, California), and the fol- −20 °C for 1 month, and then washed in water over two sieves with de- lowing thermal profile:94°Cfor15min,94°Cfor30s,57°Cfor90s, creasing meshes (0.5–0.1 mm). We identified prey species from undi- 72 °C for 60 s (40 cycles for scat, urine, and hair samples, and 35 cy- gested remains: hair, bone, hoof, and claw (medium and large-sized cles for muscle and blood samples), followed by a final extension mammals), hair and mandible (small mammals), seeds and leaves step of 72 °C for 10 min. We carried out amplifications in 10-μlvol- (fruits and plants). Moreover, hairs were washed in alcohol and ob- umes including 2 μl of DNA extraction solutions from scat, urine, served with an optical microscope (Leica DM750) to identify the species and hair samples, 1 μl from muscle or blood samples (corresponding from the characteristics of cortical scales, medulla, and root (Brunner to approximately 20–40 ng of DNA), 5 μlofQIAGENMultiplexPCR and Coman, 1974; Debrot et al., 1982; De Marinis and Asprea, 2006; Kit, 1 μl of QIAGEN Q solution (Qiagen Inc.), 0.4 μMdeoxynucleotide Teerink, 1991). We estimated the proportion of prey for each scat as triphosphates, from 0.1 to 0.4 μlof10μM primer mix (forward and they were eaten (Kruuk and Parish, 1981; Meriggi et al., 1991, 1996, reverse), and RNase-free water up to the final volume. We amplified 2015; Milanesi et al., 2012) and each prey species was assigned to a per- the mtDNA control region in a 10-μl PCR, including 1 or 2 μlofDNA cent volumetric class: b1%; 1–5%; 6–25%; 26–50%; 51–75%; 76–95%; solution, 0.3 pmol of the primers L-Pro and H350, using the following N95% that was converted in a final percent volume: 0.5%; 2.5%; 15.5%; thermal profile:94°Cfor2min,94°Cfor15s,55°Cfor15s,72°Cfor 38%; 63%; 85.6% and 98% respectively. Prey species were grouped 30 s (40 cycles), followed by a final extension of 72 °C for 5 min. PCR in six food categories (wild ungulates, livestock, small mammals, products were purified using exonuclease/shrimp alkaline phospha- medium-sized mammals, fruits, grasses). We calculated the mean per- tase (Exo-Sap; Amersham, Freiburg, Germany) and sequenced in cent volume (MV%) and the percent frequency of occurrence (FO%) both directions using the Applied Biosystems Big Dye Terminator for each food category and species of wild and domestic ungulates. kit (Applied Biosystems, Foster City, California) with the following We determined the diet composition for two main seasons (grazing steps: 96 °C for 10 s, 55 °C for 5 s, and 60 °C for 4 min of final exten- season: from April to October when livestock is on the pastures, and sion (25 cycles). non-grazing season: from November to March), for each year of study, DNA from scat, urine, and hair samples was extracted, amplified, and for each itinerary (pooling the study years), for each pack, for and genotyped in separate chambers reserved for low-template DNA pooled packs and for dispersing individuals. samples, under sterile ultraviolet laminar flow hoods, following a multiple-tube protocol (Caniglia et al., 2012, 2013), including both neg- 2.5. Data analysis ative and positive controls. We obtained genotypes from scat, urine, and hair samples replicating the analyses from 4 to 8 times, and from blood We adapted the index provided by Massolo and Meriggi (1998) as a and muscle DNA replicating the analyses twice. DNA sequences and measure of the diversity of wild ungulate community; we used the IKA microsatellites were analyzed in a 3130XL ABI automated sequencer as a measure of abundance and 5 identical classes for all wild ungulate (Applied Biosystems), using the ABI software SEQSCAPE 2.5 for se- species. We divided the range of the IKA values for all species pooled quences, GENEMAPPER 4.0 for microsatellites (Applied Biosystems) to determine the class intervals for each season because the probability (Caniglia et al., 2014) and GIMLET to control the good attribution of of detecting a track depends on the weather, mainly the presence of several samples to the same individual. snow,mud,orleaves: We assigned individual genotypes to their population of origin (wolves or dogs) using the Bayesian software STRUCTURE 2.3 (Falush Σ ¼ Ai N et al., 2003). According to previous studies (Caniglia et al., 2014), the op- DI ΣAimax K timal number of populations was set at K = 2, the value that maximized the posterior probability of the data. At K = 2, we assessed the average where A is the class of abundance of ith species, A the class of proportion of membership (qi) of the sampled populations to the in- i imax maximum abundance of the ith species, N the number of species ferred clusters. Then we assigned genotypes to the Italian wolf or dog present for a transect, and K the maximum number of species. The clusters at threshold qi = 95 (individual proportion of membership; index was calculated for each transect. Randi, 2008), or identified them as admixed if their qivalueswere We defined pack territories by Kernel Analyses at 95% on GPS intermediate. coordinates of genetic samples of wolves with parental ties. Scats within We identified familiar relationships i.e. packs, using a maximum- territories were considered to belong to the relative pack, and those out- likelihood approach (Caniglia et al., 2014) implemented in the soft- side the territories were assumed to belong to dispersing individuals. ware COLONY 2.0 (Wang and Santure, 2009). We selected all the ge- The scats localized on the overlap of two territories were not included notypes that were sampled in restricted ranges (b100 km2)atleast4 in the analysis, because their origin was not identifiable with certainty. times and for periods longer than 24 months. We determined their To estimate the minimum number of scats necessary to assess spatial distributions by 95% kernel analysis, choosing band width the diet of wolves we used the Brillouin index (1956) (Hass, 2009; using the least-squares cross-validation method (Kernohan et al., Milanesi et al., 2012; Meriggi et al., 2015): 2001; Seaman et al., 1999), using the ADEHABITATHR package for R (Calenge, 2006) and mapped them using ARCGIS 10.0. According to !−Σ ! spatial overlaps, we split individuals into distinct groups that might lnN lnni Hb ¼ correspond to packs, for which we performed parentage analyses. N We considered as candidate parents of each group all the individuals sampled in the 1st year of sampling and more than 4 times in the where Hb is the diversity of prey in the sample, N is the total number of same area and as candidate offspring all the individuals collected single prey taxa in all samples, and ni is the number of single prey taxa of within the 95% kernel spatial distribution of each pack and in a sur- the ith category. For each sample, a value of Hb was calculated and then rounding buffer area of approximately 17-km radius from the kernel re-sampled 1000 times by the bootstrap method to obtain the average (see Caniglia et al., 2014). We ran COLONY with allele frequencies values and 95% confidence intervals. We determined the adequacy of and PCR error rates as estimated from all the genotypes, assuming sample size by whether an asymptote was reached in the diversity a 0.5 probability of including fathers and mothers in the candidate curve and in the curve obtained from the incremental change in each parental pairs. Hb with the addition of two more samples. 160 C. Imbert et al. / Biological Conservation 195 (2016) 156–168
We evaluated the significance of the differences in the diet between Table 1 years and seasons by two-way non-parametric multivariate analysis of Composition of the five packs detected in Liguria region by genetic analyses from 2008 to 2013. variance (NPMANOVA), and between packs and dispersing individuals by one-way NPMANOVA with permutation (10,000 replicates), using Pack Males Females Offspring of Adopted Immigrants Migration the Bonferroni correction of the p-value for pair-wise comparisons alpha pair distance (Anderson, 2000, 2001). Annual, seasonal and pack-dispersing wolf Imperia 1 3 2 0 0 – variations of mean percent volumes of all categories and all ungulate Savona 2–32 1–21 0 – species were verified by Kruskall–Wallis test with permutation Beigua 3 2 1 2 2 97 and 75 km Antola 3 2 1 2 1 122 (10,000 replicates). Spezia 2 1 1 0 0 – Moreover we compared the observed and expected use of livestock species for each pack and for dispersing wolves by the Chi-square goodness-of-fit (Exact test) and Bonferroni's confidence interval 533 km2 for Imperia pack, 779 km2 for Savona pack, 144 km2 for Antola 2 2 analyses, testing the null hypothesis (H0) of a use proportional to the pack, 83 km for Beigua pack, and 101 km for La Spezia pack. The re- availability (Manly et al., 2002). In this analysis we considered predation maining 37 wolves apparently were not related to any pack and were cases as use; in particular we calculated the expected frequencies from considered as floating or dispersing individuals (Caniglia et al., 2014). the availability of livestock (number of heads) in each pack territory and Thirteen dispersing wolves were resampled from one to nine times outside pack areas. showing an average distance from the first to the last sampling of To identify the main factors affecting livestock consumption by 19.9 km (SE = 5.19) with a maximum distance of 60.4 km. wolves we carried out Multiple Regression Analyses (MRA) of MV% of domestic ungulates recorded for each transect vs. the variables mea- 3.2. Wolf diet sured in the buffers around the transects; only transects (N = 34) with at least 10 scats which corresponds, according to the Brillouin di- We analyzed a total of 1457 scats (year 1: 128; year 2: 276; year 3: versity index, to an incremental change of 5% for grazing season and 242; year 4: 350; year 5: 318; year 6:143) of which 863 were attributed 3% for non-grazing one, were included in the analyses. We identified to grazing season and 593 to the non-grazing one. Sample size was all the possible subsets of uncorrelated (P N 0.05) predictor variables sufficient for each year and season according to the Brillouin index by calculating the correlation matrix (Pearson product moment coeffi- (minimum sample sizes: pooled years: 23 scats; year 1: 16; year cient) among habitat variables. For each subset, we performed MRA be- 2: 19; year 3: 23; year 4: 19; year 5: 15; year 6: 15; grazing season: tween MV% of livestock and transect variables. We obtained a number 17; non-grazing season: 19). of models that were ranked by the information theoretic approach In view of the low proportion of scats attributed to dogs by genetic (Akaike, 1973). We computed the corrected value of Akaike information analyses (16 out of 389 fresh scats; 3.97%) we considered that errors criterion (AICc) because the ratio sample/parameters was small did not influence the results. By scat analyses we detected 21 kinds of (Burnham and Anderson, 2002), selecting the model with the lowest prey, pooled into eight categories (Table 2). Pooling the study years AICc as the best model and ranking the following ones by their differ- and the provinces, wild ungulates were the main food of wolves, follow- ences from the lowest AICc (Δ ). For the following analysis, we consid- i ed by domestic ungulates. Other food categories showed a mean per- ered only models with Δ ≤ 2(Best and Rayner, 2007; Burnham and i cent volume less than 3 for small mammals, medium sized mammals Anderson, 2002). Furthermore, we measured the relative importance and grasses, and less than 1% for invertebrates, fruits, and garbage of models by their Akaike weights (w ; Anderson et al., 2000, 2001). i (Fig. 2). Among wild ungulates, the most consumed were wild boar We followed the AIC statistical approach because it allows the compar- and roe deer; the others species were less used (Fig. 3A). Among live- ison of all the models, as many as the uncorrelated subsets, and the se- stock species, wolves chieflyconsumedgoats,followedbycattle(main- lection of the ones that best explain the effect of transect variables on ly calves) and sheep; horse consumption was negligible (Fig. 3B). livestock consumption. Moreover, the AIC tool allowed us to conduct an explanatory analysis taking into account all possible predictor com- binations. For each model, we calculated the Variance Inflation Factor 3.3. Temporal variations of wolf diet (VIF) in order to detect collinearity among predictor variables (Zuur et al., 2010). To validate the final model, we tested for deviation from Two-way NPMANOVA showed significant changes in wolf diet ac- normality of the residual distribution by the Shapiro–Wilk normality cording to years (F = 13.31; P b 0.0001) but not according to seasons test, for homoscedasticity by the Breush–Pagan test (Breusch and (F = 1.51; P = 0.153), and a significant interaction between the two Pagan, 1979), and for residual autocorrelation by the Durbin–Watson factors (F = 43.56; P b 0.0001). Significant differences resulted for all test (Pires and Rodrigues, 2007). pairwise comparisons with exception of year 1 versus years 2, 3, 4, and 6, between years 2 and 3, and between years 5 and 6 (Table 2). 3. Results Livestock consumption increased significantly until year 3 and then decreased. Wild ungulates increased in the diet from the first to the last 3.1. Genetic analysis study year. Small mammals showed significant variation between years with a peak in the second one and the same was for the medium-sized Genetic identifications of the 403 samples yield 205 (50.8%) reliable mammals. Also for fruits, grasses, and garbage significant but moderate multilocus genotypes, corresponding to 58 wolves (31 males M and 27 annual variations resulted (Table 2). The frequencies of occurrence of females F), 5 dogs (4 M, 1 F), 9 wolf × dog hybrid individuals (8 M, 1 F). livestock and wild ungulates were negatively correlated (Spearman
Wolf individuals were sampled from a minimum of 1 to a maximum of rank correlation: rs = −0.886; n = 6; P = 0.019) and those of 10 times. The hybrids were sampled from 1 to 3 times while the 5 dogs medium-sized and small mammals positively (rs =0.941;n=6; once each. P = 0.005). Parentage analyses led to the assigning of 20–21 wolves to five dis- Among livestock species, sheep consumption significantly decreased tinct packs (Fig. 1, Table 1), respectively named: Imperia pack, Savona in the study period, while goat and cattle increased until the third year pack (on the border between Savona and Imperia), Beigua pack (in and then decreased (Table 2). Goats strongly contributed to the overall the Mount Beigua Regional Park, on the border between Savona and livestock consumption (rs = 0.943; n = 6; P = 0.005). Concerning wild Genoa), Antola pack (in the Mount Antola Regional Park, in the province ungulate species, wild boar increased in the wolf diet until the fifth of Genoa) and Spezia pack. The minimum estimated territory sizes were study year and then decreased, roe deer increased during the study C. Imbert et al. / Biological Conservation 195 (2016) 156–168 161
Table 2 Yearly variations of mean percent volume (MV%) and frequency of occurrence (FO%) of categories and prey species in wolf diet (Liguria region, N-Italy,2008–2013).
Categories and species Year 1 Year 2 Year 3 Year 4 Year 5 Year 6
n = 128 n = 276 n = 242 n = 350 n = 318 n = 143
MV% FO% MV% FO% MV% FO% MV% FO% MV% FO% MV% FO%
Livestock 26.7 32.0 32.1 35.9 40.2 47.5 27.9 34.3 14.9 18.6 19.5 21.0 Ovisaries 5.5 19.5 4.4 13.1 4.6 15.7 3.1 12.5 2.3 13.6 0.1 3.3 Capra hircus 16.0 61.0 22.9 72.7 24.0 57.4 15.4 55.8 7.3 49.2 11.0 53.3 Bos taurus 5.3 19.5 4.4 13.1 11.7 27.0 9.1 30.8 5.3 37.3 8.4 43.3 Equus caballus 0.0 0.0 0.4 1.0 0.0 0.0 0.3 0.8 0.0 0.0 0.0 0.0 Wild ungulates 60.1 71.1 48.5 55.1 51.5 56.7 66.8 72.6 79.4 83.6 76.6 79.7 Sus scrofa 36.0 61.5 29.3 61.2 23.9 50.4 41.8 63.4 45.2 59.0 25.0 36.0 Capreolus capreolus 14.3 25.3 13.3 28.3 20.3 42.3 13.7 23.2 31.5 44.0 42.2 57.0 Cervus elaphus 4.0 7.7 1.8 3.3 4.5 8.0 5.2 7.9 0.2 0.4 0.6 0.9 Dama dama 5.9 11.0 3.8 12.5 2.9 5.1 4.4 7.5 1.2 1.5 8.1 11.4 Ovis aries musimons 0.0 0.0 0.4 0.7 0.0 0.0 0.3 0.4 0.0 0.0 0.0 0.0 Rupicapra rupicapra 0.0 0.0 0.0 0.0 0.0 0.0 1.4 2.0 1.4 1.9 0.7 0.9 Medium-sized mammals 4.4 7.0 8.0 12.7 4.4 5.4 4.0 0.9 0.6 0.9 0.7 1.4 Small mammals 2.1 2.3 5.9 7.6 1.9 5.0 0.4 0.9 0.5 0.9 1.4 1.4 Invertebrates 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 Fruits 0.1 0.8 1.5 7.6 0.8 1.7 0.9 1.1 0.9 1.6 0.0 0.0 Grasses 4.9 17.2 3.5 19.6 1.1 5.4 1.9 7.1 2.5 4.1 0.0 0.0 Garbage 0.0 0.0 0.0 0.0 0.0 0.0 0.1 1.1 0.0 0.0 0.0 0.0
NPMANOVA pairwise comparisons between years: 1–5 P = 0.003; 2–4P=0.002;2–5P=0.002;2–6 P = 0.002; 3–4 P = 0.005; 3–5P=0.002;3–6P=0.002;4–5 P = 0.005. Livestock: H = 60.26; df = 5; P b 0.0001; wild ungulates: H = 108.55; df = 5; P b 0.0001; small mammals: H = 34.77; df = 5; P b 0.0001. Medium-sized mammals: H = 68.66; df = 5, P b 0.0001; fruits: H = 40.07; df = 5, P b 0.0001; grasses: H = 77.07; df = 5; P b 0.0001. Garbage: H = 12.67; df = 5; P = 0.027. Ovis aries: H = 13.71; df = 5; P = 0.018; Capra hircus:H=46.36;df=5;Pb 0.0001; Bos taurus:H=14.63;df=5;Pb 0.012. Sus scrofa:H=49.56;df=5;Pb 0.0001; Capreolus capreolus: H = 90.20; df = 5; P b 0.0001; Cervus elaphus:H=24.59;df=5;Pb 0.0001. Dama dama: H = 20.95; df = 5; P = 0.001. period, while red and fallow deer showed significant annual variations Among livestock species, we detected significant differences for but without an evident trend (Table 2). goats and cattle, the first species being more used by Spezia pack, and For livestock species we did not find significant seasonal changes, the second by Imperia and Spezia ones. Also the use of wild ungulate whereas among wild ungulate species significant differences resulted for wild boar (H = 34.37; df = 1; P b 0.0001) and for roe deer (H = 25.50; df = 1; P b 0.0001); in particular wild boar was more consumed in non-grazing season and, on the contrary, roe deer was more used in thegrazingone(Table 3).
3.4. Variations in wolf diet between packs and dispersing wolves
Considering the five packs separately, we found overall significant differences in the use of food categories (NPMANOVA: F = 9.85; P = 0.0001); in particular the diet of Spezia pack was different from all the others. Moreover, we found significant differences comparing the diet of Beigua pack with those of Imperia and Antola. La Spezia pack consumed more livestock and medium-sized mammals and less wild ungulates and grasses than all the other packs (Table 4).
Fig. 2. Mean percent volume ± SE of prey categories in wolf diet in Liguria from 2008 Fig. 3. Mean percent volume ± SE of livestock (A) and wild ungulate species (B) in wolf to 2013. diet (Liguria region 2008–2013). 162 C. Imbert et al. / Biological Conservation 195 (2016) 156–168
Table 3 and cattle were more consumed by dispersing individuals (H = 9.17; Seasonal variations of mean percent volume (MV%) and frequency of occurrence (FO%) of df = 1; P = 0.002 and H = 7.65; df = 1; P = 0.006 respectively) but categories and prey species in wolf diet (Liguria region, N-Italy, 2008–2013). packs consumed more roe deer (H = 15.33; df = 1; P = 0.0001) and Categories and species Grazing season Non-grazing chamois (H = 5.67; df = 1; P = 0.017) (Fig. 3A and B). The diet of season dispersing wolves differed significantly from that of each pack n = 863 n = 593 (P ≤ 0.003 for all pairwise comparisons), with the exception of Spezia
MV% FO% MV% FO% pack (P = 0.623).
Livestock 28.1 32.0 25.1 29.2 Ovisaries 2.9 11.2 3.9 18.5 3.5. Livestock use versus availability Capra hircus 17.1 62.7 14.8 59.0 Bos taurus 8.1 28.6 6.2 25.4 The livestock density was greater in pack territories than in the re- Equus caballus 0.1 0.4 0.2 0.6 maining part of the study area with the exception of Antola and Savona Wild ungulates 62.7 68.7 65.9 71.0 packs (Table 5). During the study period we recorded a total of 176 pre- Sus scrofa 29.2 48.9 43.7 68.2 Capreolus capreolus 25.5 43.5 15.6 25.4 dations on livestock 15 of which upon cattle and 161 upon sheep and Cervus elaphus 3.3 5.4 2.0 3.1 goats. Considering predation cases on livestock species we found signif- Dama dama 4.2 7.8 3.4 6.2 icant differences between observed and expected frequencies of events Ovis aries musimons 0.0 0.0 0.3 0.5 across packs and dispersing wolves for cattle (χ2 = 4167.78; df = 5; Rupicapra rupicapra 0.5 0.8 1.0 1.4 b χ2 b Medium-sized mammals 3.1 4.8 2.7 4.0 P 0.0001), sheep and goats ( = 4107.74; df = 5; P 0.0001), and 2 Small mammals 2.2 3.4 1.6 2.5 for the species pooled (χ = 187.83; df = 5; P b 0.0001). In particular Invertebrates 0.0 0.2 0.0 0.0 cattle was preyed in proportion to the availability by Antola and Savona Fruits 0.6 1.9 1.1 3.2 packs, avoided by Beigua, Imperia and Spezia packs, and used more than Grasses 1.9 7.1 2.9 11.1 the availability by dispersing wolves. Sheep and goats were used as Garbage 0.02 0.2 0.01 0.3 available by Antola and Spezia packs, underused by Beigua, Imperia, and Savona packs, and overused by dispersing wolves. Pooled species species resulted different between packs; in particular wild boar were were used in proportion to the availability by Antola pack, less than consumed mainly by Imperia and Beigua packs, roe deer by Savona the availability by the other packs, and more than the availability by dis- and Beigua packs, red deer by Savona pack, fallow deer by Savona and persing wolves (Table 6). Considering the packs pooled the frequency of Antola packs, and chamois by Imperia pack (Table 4). predation events was significantly less than expected for sheep and Comparing the diets of individuals belonging to a pack and the goats (χ2 = 97.73; df = 1; P b 0.0001), and for the species pooled dispersing ones, we found overall significant differences in the use of (χ2 = 103.64; df = 1; P b 0.0001) but for cattle (χ2 = 4.29; df = 1; prey categories (NPMANOVA: F = 32.24; P b 0.0001). In particular, a P = 0.066); pooled packs underused cattle, sheep and goats, and the higher consumption of livestock (H = 29.44; df = 1; P b 0.0001) and pooled species (Table 6). medium-sized mammals (H = 10.98; df = 1; P = 0.001) resulted for dispersing wolves, whereas the contrary was the case for wild ungulate 3.6. Model of livestock consumption use, higher in pack diet than in that of dispersing individuals (H = 40.01; df = 1; P b 0.0001) (Fig. 2). Significant differences resulted also By the Multiple Linear Regression Analyses on the subsets of uncor- considering livestock and wild ungulate species. In particular goats related predictors, we obtained only one model, the others having
Table 4 Mean percent volume (MV%) and frequency of occurrence (FO%) of categories and prey species in the diet of wolf packs (Liguria region, N-Italy, 2008–2013).
Categories and species Imperia pack Savona pack Beigua pack Antola pack La Spezia pack
n = 297 n = 102 n = 213 n = 137 n = 64
MV% FO% MV% FO% MV% FO% MV% FO% MV% FO%
Livestock 24.2 27.9 17.6 20.6 13.1 14.6 18.2 27.0 46.0 50.0 Ovis aries 2.6 14.5 1.9 9.5 2.9 22.6 1.7 16.2 2.9 6.3 Capra hircus 11.7 53.0 13.1 76.2 8.2 64.5 12.9 70.3 37.2 81.3 Bos taurus 9.7 42.2 2.6 14.3 2.0 19.4 3.6 16.2 5.9 12.5 Equus caballus 0.3 1.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Wild ungulates 69.7 73.1 75.0 83.3 81.3 84.5 64.6 78.1 44.4 50.0 Sus scrofa 45.3 66.4 19.2 27.1 38.7 51.7 30.0 47.7 31.6 75.0 Capreolus capreolus 19.8 29.5 35.1 48.2 35.2 48.9 19.3 33.6 11.3 25.0 Cervus elaphus 1.6 2.3 9.8 12.9 3.6 4.4 1.9 2.8 0.0 0.0 Dama dama 0.0 0.0 10.0 11.8 3.8 5.6 13.4 24.3 0.0 0.0 Ovis aries musimons 0.3 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Rupicapra rupicapra 2.6 3.7 1.0 1.2 0.0 0.0 0.0 0.0 1.5 0.0 Medium-sized mammals 1.0 1.0 0.4 1.0 1.0 1.4 3.1 8.0 7.7 7.8 Small mammals 2.2 3.0 3.3 0.0 0.9 0.9 4.3 5.1 0.0 7.8 Invertebrates 0.002 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Fruits 0.1 0.3 1.4 3.9 0.0 0.0 2.9 8.0 0.2 1.6 Grasses 0.4 5.4 2.4 4.9 1.9 7.0 7.9 19.7 0.3 4.7 Garbage 0.02 0.7 0.2 2.0 0.0 0.0 0.0 0.0 0.0 0.0
NPMANOVA pairwise comparisons between packs: Imperia–Beigua P = 0.008; Imperia–La Spezia P = 0.001; Savona La Spezia P = 0.001; Beigua–Antola P = 0.01; Beigua–La Spezia P = 0.001; Antola–La Spezia P = 0.001. Livestock: H = 36.31; df = 4; P b 0.0001; wild ungulates: H = 45.59; df = 4; P b 0.0001; medium-sized mammals: H = 24.97; df = 4; P b 0.0001. Fruits: H = 34.10; df = 4; P b 0.0001; grasses: H = 31.59; df = 4; P b 0.0001. Capra hircus:H=38.08;df=4;Pb 0.0001; Bos taurus:H=20.54;df=4;Pb 0.0001. Sus scrofa:H=27.26;df=4;Pb 0.0001; Capreolus capreolus: H = 34.92; df = 4; P b 0.0001; Cervus elaphus:H=24.17;df=4;Pb 0.0001. Dama dama:H=69.57;df=4;Pb 0.0001; Rupicapra rupicapra:H=9.53;df=4;P=0.049. C. Imbert et al. / Biological Conservation 195 (2016) 156–168 163
Table 5 Table 7 Percentage of pastures and density (heads per km2) of livestock species in pack territories Results of multiple regression analysis of mean percent volume of domestic ungulates in and in non-pack area. the wolf diet vs. the transect variables (N = 34).
Pack Pastures (%) Cattle Sheep and goats Total Transect variables Regression Standardized t P VIF coefficients (SE) coefficients Antola 7.6 1.4 0.1 1.5 Beigua 4.7 1.9 3.3 5.2 Intercept 57.2 (6.16) 9.29 b0.0001 Imperia 8.5 3.7 14.1 17.8 Pack presence −31.4 (3.28) −0.85 9.59 b0.0001 1.3 Savona 1.8 0.5 0.5 1.0 Pasture number 0.9 (0.15) 0.56 6.45 b0.0001 1.2 Spezia 8.6 7. 5 2.5 10.0 Prevention (%) −37.5 (7.08) −0.45 5.29 b0.0001 1.1 Pooled packs 4.7 2.1 5.1 7.3 Roe deer abundance (IKA) −15.7 (3.92) −0.36 3.99 0.001 1.3 No pack 2.6 0.8 1.7 2.5 Deciduous woods (%) −0.4 (0.12) −0.29 3.48 0.002 1.1 Wild Ungulate Diversity 9.8 (4.27) 0.21 2.29 0.031 1.4 Index
R2 = 0.807. SEE = 7.68. F = 22.59. df = 6,25. P b 0.0001. ΔAICc N 2(Table 7). Six variables with significant regression coefficients entered the model explaining 80.7% of the variance of the mean percent volume of livestock in the wolf diet. The presence of packs, prevention methods, deciduous woods and roe deer abundance had a negative ef- et al., 2011; Migli et al., 2005; Papageorgiou et al., 1994; Peterson and fect on livestock consumption, whereas the number of pastures in the Ciucci, 2003; Vos, 2000; Zlatanova et al., 2014). Usually the former are areas covered by transects and the diversity of wild ungulate communi- found in areas where there are rich and abundant wild ungulate guilds ty had a positive effect (Fig. 4). The presence of a pack (by opposition to and where livestock is inaccessible because of the husbandry methods, dispersing wolves) had the strongest influence followed by the number and the latter where wild ungulates are rare and livestock is free ranging of pastures, the percentage of pastures with prevention methods, roe and unguarded (Cuesta et al., 1991; Meriggi and Lovari, 1996; Okarma, deer abundance, percentage of deciduous forest, and finally wild 1995; Peterson and Ciucci, 2003; Zlatanova et al., 2014). Considering ungulate diversity index (Table 7). The Variance Inflation Factor re- Europe as a whole, the importance of wild ungulates in the wolf diet vealed no collinearity among predictors, and the residuals of the seems to follow a cline decreasing from North to South and an increas- model were normally distributed (Shapiro–Wilk test = 0.98, P = ing trend in particular after the eighties in the last century (Meriggi and 0.736), not auto-correlated (Durbin–Watson statistic = 2.30) and the Lovari, 1996; Meriggi et al., 2011; Okarma, 1995; Zlatanova et al., 2014). homoscedasticity was respected (Breush–Pagan statistic = 6.32; df = Wolves in the Liguria region use fewer wild ungulates and more live- 6; P = 0.389). Examining the relationships between predictors, stock than those of other close areas located in the northern Apennines transects with wolf packs were characterized by a greater presence of (Capitani et al., 2004; Mattioli et al., 1995, 2004, 2011; Meriggi et al., pastures, percentage of deciduous woods, and roe deer abundance in 1996, 2011, 2015; Milanesi et al., 2012). These differences can be related respect to those with dispersing wolves, whereas the percentage of to the characteristics of wild prey community in Liguria where there are livestock farms with prevention methods and the wild ungulate two widespread species locally very abundant (wild boar and roe deer), diversity index were lower (Fig. 5). However these differences were other two localized but with high density populations (fallow deer and not significant (Mann–Whitney U test, P N 0.05 in all cases). chamois) and the last two (red deer and mouflon) are at present rare. Moreover few livestock farms adopt prevention methods, leaving herds, in particular goat flocks, free ranging and unguarded on pastures 4. Discussion during the grazing season. This situation could also cause the annual variations of the use of The diet of wolves in the Liguria region is characterized by a wild ungulates and livestock; both being probably related to the fluctu- medium–high occurrence of wild ungulates and by an important part ations in abundance of the main wild prey species (wild boar and roe consisting of large domestic prey, the other food categories being a deer) because of the quite constant number of livestock heads reared negligible fraction of the diet. This picture places the food habits of in the Liguria region. The close negative relationships between the im- wolves in our study area between those of populations preying almost portance of the two main food categories in the wolf diet over the exclusively on wild herbivores and those of wolves living mainly at study period demonstrates that livestock is more used when wild herbi- the expense of livestock and other food of human origin, that can be vores are less available (Meriggi and Lovari, 1996; Meriggi et al., 2011). found in human altered landscapes of some South and East European In any case, wild herbivores showed an increase in the wolf diet during countries such as Portugal, Spain, South-central Italy, Greece, Bulgaria, the study period in accordance with the general trend already found in Ukraine, Moldova, and Azerbaijzan (Meriggi and Lovari, 1996; Meriggi Europe and in particular in Italy, and in agreement with the ability of wolves to respond in a short time to the changes in abundance of the preferred prey species (Meriggi and Lovari, 1996; Meriggi et al., 2011, 2015; Peterson and Ciucci, 2003). Table 6 Wolves in Liguria consumed mainly wild boar, the main wild prey in Results of Bonferroni simultaneous confidence interval analyses for the differences be- tween expected (EUP) and observed usage proportion (OUP) of livestock species across the Mediterranean range of the wolf (Meriggi and Lovari, 1996; Meriggi packs and dispersing wolves. et al., 2011; Okarma, 1995; Zlatanova et al., 2014). This choice could be due to the high wild boar abundance, and to the fact that the species Pack Cattle (n = 15) Sheep and goats Pooled species lives in large groups easily detectable by a predator. Furthermore, births (n = 161) (n = 176) occur all year round, causing the removal from the matriarchal groups EUP OUP EUP OUP EUP OUP of sub-adults; these individuals are profitable prey because they have Antola 0.030 0.133 0.001 0.012 0.010 0.023 the body size of an adult without its experience so that the handling a a a Beigua 0.024 0.000 0.019 0.000 0.020 0.000 time can be minimized (Meriggi et al., 1996, 2011, 2015; Milanesi Imperia 0.302 0.000a 0.511 0.019a 0.447 0.017a Savona 0.060 0.067 0.026 0.000a 0.037 0.006a et al., 2012). Spezia 0.116 0.000a 0.017 0.012 0.048 0.011a The second wild ungulate in order of importance was the roe deer; Pooled packs 0.532 0.200a 0.575 0.043a 0.562 0.057a roe deer has solitary behavior so its detectability and its encounter a a a No pack 0.468 0.800 0.425 0.957 0.438 0.943 rate are low compared to the wild boar, with the exception of the a Differences at minimum level of α =0.05. areas where the species is present with high density (Meriggi et al., 164 C. Imbert et al. / Biological Conservation 195 (2016) 156–168
Fig. 4. Relationships between mean volume (%) of livestock in wolf diet and the transect variables entered in the regression model.
2011, 2015; Milanesi et al., 2012). Among the other wild ungulate spe- the latter is more difficult to prey upon because of the low accessibility cies only fallow deer reached a limited importance in the last study of the habitats (Meriggi and Lovari, 1996; Patalano and Lovari, 1993; year; this species, together with the chamois, is locally abundant but Poulle et al., 1997). Red deer and mouflon are used only occasionally
Fig. 5. Average values (±SE) of transect variables entered the model of livestock consumption in relation to the pack presence. C. Imbert et al. / Biological Conservation 195 (2016) 156–168 165 because of their rarity. Despite the annual variations of the wild ungu- selection for livestock species by dispersing wolves whereas packs late species in the wolf diet, only for roe deer did we find a trend with underuse or use as availability both cattle or sheep and goats or the an increase of three times the importance from the first to the last species pooled. study year. This is in accordance with the general increase of the Italian The model of livestock consumption explained a high percentage of population of roe deer in the last decades (Carnevali et al., 2009). the variance and it was therefore very informative. The presence of The importance of livestock species in the wolf diet in Liguria region packs, unlike the case of dispersing individuals, had a negative effect was not in agreement with their respective abundance; indeed, the on livestock consumption. This is in accord with the lower use of main prey species were goats and cattle that are respectively the third livestock species that we found in the pack diet compared to that of and the second in number. Goats are particularly vulnerable to wolf pre- dispersing wolves; structured packs hunt on their territory and know dation because they are left unguarded and free ranging on mountains; where to find wild prey, whereas dispersing individuals, new to the moreover goats lost in the mountain can survive, forming groups of feral area, do not know it and hunt the first potential prey they encounter. animals, available all year round for wolves, and these groups of feral The number of pastures had a positive effect on consumption of goats are increasing in number and size in Liguria. As far as cattle are livestock because if the pastures are fragmented and scattered in the concerned, wolves prey almost exclusively upon calves born during forests, the contact zone between woods and pastures increases and the grazing period on pastures, whereas adult cows are rarely attacked; this can enhance the predation risk facilitating the attacks by wolves so only cattle farms that adopt calf births on pastures are vulnerable (Dondina et al., 2014; Kaartinen et al., 2009; van Lière et al., 2013). (Brangi et al., 1992; Meriggi et al., 1991, 1996). Prevention methods negatively affected the livestock consumption; Surprisingly, we did not find significant changes of the food catego- the effect of the adoption of different methods (nocturnal shelter, pres- ries from grazing to non-grazing seasons, with the exception of fruits ence of shepherds and dogs, electric fences) of herd and flock protection that were more eaten in the grazing one. As for wild ungulates, wolves in reducing predator attacks and their success was demonstrated by used wild boar in winter and roe deer in summer; in summer, roe deer several studies even if in some cases they fail or are impossible to are more vulnerable because of the presence of young, and in winter, adopt (Dondina et al., 2014; Espuno et al., 2004; Landry et al., 1999; several wild boars are injured and not retrieved by hunters during Mech and Boitani, 2003; Miller, 2015). Roe deer abundance also drive hunts and consequently are easily found and caught by wolves. decreased the use of livestock; the presence of roe deer corresponds Moreover snow depth makes wild boars more vulnerable to predation to the presence of a second potential prey species for wolves, the first (Okarma, 1995). As for livestock, we found that wolves eat it, particular- one being wild boar that is present in the whole region at high density. ly goats, also in winter, although in this season they should be in the In this situation, if the abundance of one of the two species drops, sheepfolds. This pattern was also found by Patalano and Lovari (1993) wolves can compensate with the other one to satisfy their food require- in the Abruzzo National Park (central Italy). Two reasons can be pro- ments and thus neglect livestock. In particular, wild boar populations posed: firstly, the scat analysis does not permit the making of distinc- are subject to substantial fluctuations related to the occurrence of tions between consumption of preyed animals and of carcasses, thus mast seeding years and the presence of a secondary prey at medium– wolves can feed on carcasses of lost animals during winter that have high density can stabilize the overall availability of prey species been well conserved by snow; also shepherds sometime leave their live- (Bieber and Ruf, 2005). This is in accord with Meriggi and Lovari stock on pastures for a part of winter, exposing it to wolf attacks during (1996) and with Meriggi et al. (2011) which suggest the increase of the cold season. the diversity of wild ungulate community as a measure to mitigate the The highlighted differences of the diets between packs seem to be conflicts with husbandry. The extent of deciduous woods decreased partially related to the local variation of wild ungulate species, and con- livestock consumption, probably in relation to the great density of sequently to the composition of the wild ungulate community. In partic- wild ungulates (wild boar and deer) that can be find in this kind of veg- ular, the packs located in the provinces with the highest density of roe etation (Focardi et al., 2009; Fonseca, 2008); in fact, usually the presence deer (Savona and the western part of Genoa, respectively 38.6 and of large wild herbivores decreases predation on livestock by wolves 46.7 individuals per km2) consumed more roe deer than the other (Meriggi and Lovari, 1996; Meriggi et al., 1996, 2011). Surprisingly, packs; moreover, fallow deer and chamois occurred almost exclusively the diversity of wild ungulate community had a positive effect on live- in the diet of packs living in the areas where these species are present stock use, a rich and abundant community of wild ungulates reducing or abundant. Moreover, Spezia pack has a diet with a high percentage the consumption of livestock elsewhere (Meriggi et al., 1996; Meriggi of livestock compared to the other packs. In this case, pack adaptation and Lovari, 1996). In our case the wild ungulate diversity index was pos- to feeding on livestock could be the result of the scarcity of wild prey itively related to the presence of chamois; this species is very localized (Meriggi and Lovari, 1996; Meriggi et al., 2011; Vos, 2000). However and abundant only in the Imperia province where the roe deer density packs do not hunt only according to prey abundance, but accessibility, and wild boar abundance are lower than in other parts of Ligurian vulnerability and profitability of prey as well as composition of ungulate region. If wolves have the choice between chamois and easier prey, community, wolf foraging behavior, previous hunting experience, cul- e.g. livestock, they will prefer the latter. tural transmission, and learning from parents can heavily affect food choice and predator diet (Curio, 1976; Endler, 1991; Huggard, 1993; Meriggi et al., 1996). 5. Conclusion We also found that the packs consume more wild ungulates than the dispersing wolves, and dispersing individuals showed a greater use of In Liguria, as in many countries of southern Europe, conflicts be- livestock than packs. Dispersing wolves are mainly young individuals tween wolf conservation and husbandry are far from being solved and and their hunting success is usually lower than that of older ones; be- they are an important threat to wolf conservation, as the high number cause of this they could direct predation on livestock that, because of of wolves found illegally killed demonstrates (12 individuals out of 16 domestication, have less effective defenses against predators than wild confirmed dead between 2007 and 2014 in the whole region). Poaching large prey (Meriggi and Lovari, 1996; Meriggi et al., 1996). Moreover, by shooting and poisoning is the main mortality factor of wolves in the dispersing wolves can cover great distances in a short time and region and in Italy, and can be related to the damage to livestock farms therefore do not have the time to learn the wild prey distribution (Lovari et al., 2007). Consequently it is important for wolf conservation (Linnell et al., 1999); as a consequence dispersing individuals can at- to adopt management options that can effectively protect Ligurian tack livestock herds that have a greater detectability because of their wolves, to maintain a connection between sub-populations of Alps highly clumped distribution and their small movement. This finding and Apennines, avoiding the isolation of Alpine wolves, and to permit is confirmed by the use vs. availability analyses that showed the linkup between Italian and Balkan populations. 166 C. Imbert et al. / Biological Conservation 195 (2016) 156–168
The main results of our study useful to improving wolf conservation Appendix A. Environmental variables measured in the and planning effective management actions aimed at conflict mitigation 13-km buffers around the transects and used to model are as follows: i) the relationship between livestock and wild ungulate livestock consumption consumption, ii) the differences in livestock and wild ungulate use be- tween packs, iii) the differences in diet between packs and dispersing Environmental variables Unit wolves, and iv) the model of livestock consumption showing that the Mixed woods % main factors negatively affecting predation upon livestock species are Deciduous woods % the presence of packs, the adoption of prevention methods, and roe Coniferous woods % deer abundance. Scrublands % In order to limit the damage that wolves cause on husbandry, con- Pastures % Rocky areas % servation measures should be primarily aimed at restoring a rich and Water (streams and lakes) % abundant wild ungulate community. This goal can be achieved by a bet- Urban areas % ter regulation of wild boar and roe deer hunting and by more effective Altitude Meters a.s.l. harvest plans in order to maintain stable the population of the former, Slope Degree Annual insolation kW/m2 and to increase the density of the latter, in particular in those areas where it is at low density. Moreover, reintroductions of red deer should be carried out to increase its presence in the region. References Another important step for conflict mitigation is to encourage the presence of wolf packs that at present are limited in number in re- Akaike, H., 1973. Maximum likelihood identification of Gaussian autoregressive moving average models. Biometrika 60, 255–265. spect to the availability of suitable areas in the region (Meriggi Anderson, M.J., 2000. NPMANOVA: A FORTRAN Computer Program for Non-parametric et al., 2013, unpublished report). If all available and suitable areas Multivariate Analysis of Variance (for any two-factor ANOVA design) Using Permuta- were occupied by packs the presence of erratic wolves would be re- tion Tests. Department of Statistics, University of Auckland, Auckland, New Zealand. Anderson, M.J., 2001. A new method for non-parametric multivariate analysis of variance. duced because of the intolerance of packs members towards foreign Austral Ecol. 26, 32–46. individuals (Mech, 1970; Mech and Boitani, 2003), and livestock Anderson, D.R., Burnham, K.P., Thompson, W.L., 2000. Null hypothesis testing: problems, depredation lowered. To enhance the pack numbers in Liguria the prevalence and an alternative. J. Wildl. Manag. 64, 912–923. main action is poaching repression; illegal kills can cause pack Anderson, D.R., Link, W.A., Johnson, D.H., Burnham, K.P., 2001. Suggestion for presenting the results of data analyses.J. Wildl. Manag. 65, 373–378. break up and social disruption with an increase of dispersal and the Ansorge, H., Kluth, G., Hahne, S., 2006. Feeding ecology of wolves Canis lupus returning to formation of new breeding pairs in other areas, the ultimate effect Germany. Acta Theriol. 51, 99–106. of this being a low effectiveness of wild prey use and a consequent Apollonio, M., Mattioli, L., Scandura, M., Mauri, L., Gazzola, A., Avanzinelli, E., 2004. Wolves in the Casentinesi forests: insights for wolf conservation in Italy from a protected area increase of livestock depredation (Haber, 1996; Sand et al., 2006; with a rich wild prey community. Biol. Conserv. 120, 249–260. Wielgus and Peebles, 2014). Ausband, D.E., Stansbury, C.R., Stenglein, J.L., Struthers, J.L., Waits, L.P., 2015. Recruitment Prevention methods are important in reducing livestock con- in a social carnivore before and after harvest. Anim. Conserv. 18, 415–423. Balciauskas, L., 2008. Wolf numbers and distribution in Lithuania and problems of species sumption but they are not applicable everywhere, in particular conservation. Ann. Zool. Fenn. 45, 329–334. on very large pastures and because of the increased costs of Barabesi, L., Fattorini, L., 2013. Random versus stratified location of transects or points in breeding. distance sampling: theoretical results and practical considerations. Environ. Ecol. Stat. 20, 215–236. On the basis of our results numeric control seems to be question- Barabesi, L., Franceschi, S., 2011. Sampling properties of spatial total estimators under tes- able. In a pack, removal of one of the two alpha members can lead to sellation stratified designs. Environmetrics 22, 271–278. its break up and the other individuals leave the territory (Mech and Behdarvand, N., Kaboli, M., Ahmadi, M., Nourani, E., Mahini, A.S., Aghbolaghi, M.A., 2014. Spatial risk model and mitigation implications for wolf–human conflict in a highly Boitani, 2003). Consequently, livestock attacks can decrease drasti- modified agroecosystem in western Iran. Biol. Conserv. 177, 156–164. cally, the wolves not being any longer present in the area. However, Best, D.J., Rayner, J.C., 2007. Chi-squared components for tests of fit and improved models it is a brief effect because empty suitable areas are rapidly re- for the grouped exponential distribution. Comput. Stat. Data Anal. 51, 3946–3954. colonized by dispersing individuals, who have a bigger consumption Bieber, C., Ruf, T., 2005. Population dynamics in wild boar Sus scrofa: ecology, elasticity of growth rate and implications for the management of pulsed resource consumers. of livestock than packs; these dispersing individuals should quickly J. Appl. Ecol. 42, 1203–1213. change into residents and form a pack but this process requires sev- Boitani, L., 2000. Action plan for the conservation of wolves in Europe (Canis lupus). eral years during which livestock depredation increases. So removal Nature and Environment, No. 113. Council of Europe Publishing. Brangi, A., Rosa, P., Meriggi, A., 1992. Predation by wolves (Canis lupus) on wild and do- measures do not solve the problem in the long run but conversely mestic ungulates in northern Italy. In: Spitz, F., Janeau, G., Gonzales, G., Alaignier, S. they can amplify it also putting at risk the wolf population because of (Eds.), International Symposium “Ongulés/Ungulates 91”. Toulouse, pp. 541–543. the direct and indirect effects of harvest on recruitment (Ausband Breitenmoser, U., 1998. Large predators in the Alps: the fall and rise of man's competitors. Biol. Conserv. 83, 279–289. et al., 2015; Wielgus and Peebles, 2014). Use of prevention tools, pro- Breusch, T.S., Pagan, A.R., 1979. A simple test for heteroscedasticity and random coeffi- motion of a rich wild ungulate community and avoiding numerical con- cient variation. Econometrica 47, 1287–1294. trol and poaching must be used together to mitigate conflicts between Brillouin, L., 1956. Science and Information Theory. Academic Press, NewYork, USA. Brunner, H., Coman, B.J., 1974. The Identification of Mammalian Hair. Inkata Press, wolf conservation and husbandry. They have to be combined with Melbourne, Australia. good monitoring of wolf populations, so that which stage of coloniza- Burnham, K.P., Anderson, D.R., 2002. Model Selection and Multimodel Inference: A Prac- tion wolves are at can be known. tical Information-Theoretic Approach. second ed. Springer-Verlag, New York, USA. Calenge, C., 2006. The package “adehabitat” for the R software: a tool for the analysis of space and habitat use by animals. Ecol. Model. 197, 516–519. Caniglia, R., Fabbri, E., Cubaynes, S., Gimenez, O., Lebreton, J.D., Randi, E., 2012. An im- Acknowledgments proved procedure to estimate wolf abundance using non-invasive genetic sampling and capture–recapture mixture models. Conserv. Genet. 13, 53–64. Caniglia, R., Fabbri, E., Mastrogiuseppe, L., Randi, E., 2013. Who is who? Identification of This research was supported by the Liguria Region, by the Regional livestock predators using forensic genetic approaches. Genetics 7, 397–404. Park “M. Antola” (Regional Operational Program — European Regional Caniglia, R., Fabbri, E., Galaverni, M., Milanesi, P., Randi, E., 2014. Noninvasive sampling Development Fund 2007–2013 - CUP B37E12000040009). We thank F. and genetic variability, pack structure, and dynamics in an expanding wolf popula- tion. J. Mammal. 95, 41–59. Puopolo, L. Schenone, D. Signorelli, and M. Zanzottera for their help in Capitani, C., Bertelli, I., Varuzza, P., Scandura, M., Apollonio, M., 2004. Acomparative field work as well as the personnel of the Forestry Corp and Provincial analysis of wolf (Canis lupus) diet in three different Italian ecosystems. Mamm. Biol. Police who helped in collecting data. V. Devictor, S. Lovari, and an anon- 69, 1–10. Carnevali, L., Pedrotti, L., Riga, F., Toso, S., 2009. Banca dati Ungulati. Status, ymous referee provided helpful comments to an early draft of the distribuzione,consistenza, gestione e prelievo venatorio delle popolazioni di Ungulati manuscript. in Italia. Rapporto 2001–2005. Biol. Conserv. Fauna 117, 1–168. C. Imbert et al. / Biological Conservation 195 (2016) 156–168 167
Chapron, G., Legendre, S., Ferrière, R., Clobert, J., Haight, R.G., 2003. Conservation and con- Marucco, F., McIntire, E.J.B., 2010. Predicting spatio-temporal recolonization of large trol strategies for the wolf (Canis lupus) in Western Europe based on demographic carnivore populations and livestock depredation risk: wolves in the Italian Alps. models.C.R.Biol.326,575–587. J. Appl. Ecol. 47, 789–798. Chapron, G., Kaczensky, P., Linnell, J.D., Von Arx, M., Huber, D., Andrén, H., ... Boitani, L., Massolo, A., Meriggi, A., 1998. Factors affecting habitat occupancy by wolves in northern 2014. Recovery of large carnivores in Europe's modern human-dominated land- Apennines (northern Italy): a model of habitat suitability. Ecography 21, 97–107. scapes. Science 346, 1517–1519. Mattioli, L., Apollonio, M., Mazzarone, V., Centofanti, E., 1995. Wolf food habits and wild Ciucci, P., Boitani, L., Francisci, F., Andreoli, G., 1997. Home range, activity and movements ungulate availability in the Foreste Casentinesi National Park, Italy. Acta Theriol. 40, of a wolf pack in central Italy. J. Zool. 243, 803–819. 387–402. Corsi, F., Duprè, E., Boitani, L., 1999. A large-scale model of wolf distribution in Italy for Mattioli, L., Capitani, C., Avanzinelli, E., Bertelli, I., Gazzola, A., Apollonio, M., 2004. conservation planning. Conserv. Biol. 13, 150–159. Predation by wolves (Canis lupus) on roe deer (Capreolus capreolus) in north- Cuesta, L., Barcena, F., Palacios, F., Reig, S., 1991. The trophic ecology of the Iberian Wolf eastern Apennine, Italy. J. Zool. 264, 249–258. (Canis lupus signatus Cabrera,1907) — a new analysis of stomach's data. Mammalia Mattioli, L., Capitani, C., Gazzola, A., Scandura, M., Apollonio, M., 2011. Prey selection and 55, 239–255. dietary response by wolves in a high-density multi-species ungulate community. Eur. Curio, E., 1976. The Ethology of Predation. Springer, Berlin, Germany. J. Wildl. Res. 57, 909–922. De Marinis, A.M., Asprea, A., 2006. Hair identification key of wild and domestic ungulates Mech, L.D., 1970. The Wolf: The Ecology and Behavior of an Endangered Species. The from southern Europe. Wildl. Biol. 12, 305–320. Natural History Press, New York, USA. Debrot, S., Mermod, C., Fivaz, G., Weber, J.M., 1982. Atlas des poils de mammifères Mech, L.D., Boitani, L., 2003. Wolf social ecology. In: Mech, L.D., Boitani, L. (Eds.), Wolves: d'Europe. Insitut de Zoologie de l'Université de Neuchâtel, Neuchâtel, Swiss. Behavior, Ecology, and Conservation. University of Chicago Press, Chicago, USA, Dondina, O., Meriggi, A., Dagradi, V., Perversi, M., Milanesi, P., 2014. Wolf predation on pp. 1–34. livestock in an area of northern Italy and prediction of damage risk. Ethol. Ecol. MEEDDAT–MAP, 2008. Plan d'action national sur le loup 2008–2012, dans le contexte Evol. 27, 200–219. français d'une activité importante et traditionnelle d'élevage. Ministère de l'Ecologie, Eggermann, J., da Costa, G.F., Guerra, A.M., Kirchner, W.H., Petrucci-Fonseca, F., 2011. de l'Energie, du Développement Durable et de l'Aménagement du Territoire et Presence of Iberian wolf (Canis lupus signatus) in relation to land cover, livestock Ministère de l'Agriculture et de la Pêche, Paris, France. and human influence in Portugal. Mamm. Biol. 76, 217–221. Meriggi, A., Lovari, S., 1996. A review of wolf predation in southern Europe: does the wolf Endler, J.A., 1991. Interactions between predators and prey. In: Krebs, J.R., Davies, N.B. prefer wild prey to livestock? J. Appl. Ecol. 1561–1571. (Eds.), Behavioural Ecology. An Evolutionary Approach. Blackwell Sci. Publ., Oxford, Meriggi, A., Rosa, P., Brangi, A., Matteucci, C., 1991. Habitat use and diet of the wolf in pp. 169–196. northern Italy. Acta Theriol. 36, 141–151. Espuno, N., Lequette, B., Poulle, M.L., Migot, P., Lebreton, J.D., 2004. Heterogeneous re- Meriggi, A., Brangi, A., Matteucci, C., Sacchi, O., 1996. The feeding habits of wolves in rela- sponse to preventive sheep husbandry during wolf recolonization of the French tion to large prey availability in northern Italy. Ecography 19, 287–295. Alps. Wildl. Soc. Bull. 32, 1195–1208. Meriggi, A., Brangi, A., Schenone, L., Signorelli, D., Milanesi, P., 2011. Changes of wolf Fabbri, E., Miquel, C., Lucchini, V., Santini, A., Caniglia, R., Duchamp, C., Weber, J.M., (Canis lupus) diet in Italy in relation to the increase of wild ungulate abundance. Lequette, B., Marucco, F., Boitani, L., Fumagalli, L., Taberlet, P., Randi, E., 2007. From Ethol. Ecol. Evol. 23, 195–210. the Apennines to the Alps: colonization genetics of the naturally expanding Italian Meriggi, A., Dagradi, V., Dondina, O., Perversi, M., Milanesi, P., Lombardini, M., Raviglione, wolf (Canis lupus) population. Mol. Ecol. 16, 1661–1671. S., Repossi, A., 2015. Short term responses of wolves feeding habits to changes of wild Fabbri, E., Caniglia, R., Kusak, J., Galov, A., Gomercic, T., Arbanasic, H., Huber, D., Randi, E., and domestic ungulate abundance in Northern Italy. Ethol. Ecol. Evol. 27, 389–411. 2014. Genetic structure of expanding wolf (Canis lupus) populations in Italy and Migli, D., Youlatos, D., Iliopoulos, Y., 2005. Winter food habits of wolves in Central Greece. Croatia, and the early steps of the recolonization of the Eastern Alps. Mamm. Biol. J. Biol. Res. (Thessaloniki) 4, 217–220. 79, 138–148. Milanesi, P., Meriggi, A., Merli, E., 2012. Selection of wild ungulates by wolves Canis lupus Falush, D., Stephens, M., Pritchard, J.K., 2003. Inference of population structure from (L. 1758) in an area of the Northern Apennines (North Italy). Ethol. Ecol. Evol. 24, multilocus genotype data: linked loci and correlated allele frequencies. Genetics 81–96. 164, 1567–1587. Miller, J.R.B., 2015. Mapping attack hotspots to mitigate human–carnivore conflict: ap- Focardi, S., Montanaro, P., La Morgia, V., Riga, F., 2009. Piano d'azione nazionale per il proaches and applications of spatial predation risk modeling. Biodivers. Conserv. capriolo italico (Capreolus capreolus italicus). Quad. Cons. Nat. 31, 1–172. 24, 2887–2911. Fonseca, C., 2008. Winter habitat selection by wild boar Sus scrofa in southeastern Poland. Mills, L.S., Citta, J.J., Lair, K.P., Schwartz, M.K., Tallmon, D.A., 2000. Estimating animal abun- Eur. J. Wildl. Res. 54, 361–366. dance using noninvasive DNA sampling: promise and pitfalls. Ecol. Appl. 10, 283–294. Fritts, S.H., Mech, L.D., 1981. Dynamics, movements, and feeding ecology of a newly Okarma, H., 1995. The trophic ecology of wolves and their predatory role in ungulate protected wolf population in northwestern Minnesota. Wildl. Monogr. 80, 3–79. communities of forest ecosystems in Europe. Acta Theriol. 40, 335–386. Genovesi, P., 2002. Piano d'azione nazionale per la conservazione del lupo (Canis lupus). Papageorgiou, N., Vlachos, C., Sfougaris, A., Tsachalidis, E., 1994. Status and diet of wolves Quad. Cons. Nat. 13, 1–94. in Greece. Acta Theriol. 39, 411–416. Haber, G.C., 1996. Biological, conservation, and ethical implications of exploiting and Patalano, M., Lovari, S., 1993. Food habits and trophic niche overlap of the wolf Canis controlling wolves. Conserv. Biol. 10, 1068–1081. lupus, (L. 1758), and the red fox Vulpes vulpes (L. 1758) in a Mediterranean mountain Hass, C.C., 2009. Competition and coexistence in sympatric bobcats and pumas. J. Zool. area. Rev. Ecol. (Terre Vie) 48, 279–294. 278, 174–180. Peterson, R.O., Ciucci, P., 2003. The wolf as a carnivore. In: Mech, L.D., Boitani, L. (Eds.), Huggard, D.J., 1993. Prey selectivity of wolves in Banff National Park. I. Prey species. Can. Wolves: Behavior, Ecology, and Conservation. University of Chicago Press, Chicago, J. Zool. 71, 130–139. USA, pp. 104–130. Iacolina, L., Scandura, M., Gazzola, A., Cappai, N., Capitani, C., Mattioli, L., Vercillo, F., Pires, A.M., Rodrigues, I.M., 2007. Multiple linear regression with some correlated errors: Apollonio, M., 2010. Y-chromosome microsatellite variation in Italian wolves: a classical and robust methods. Stat. Med. 26, 2901–2918. contribution to the study of wolf–dog hybridization patterns. Mamm. Biol. 75, Poulle, M.L., Carles, L., Lequette, B., 1997. Significance of ungulates in the diet of recently 341–347. settled wolves in the Mercantour Mountains (South Eastern France). Rev. Ecol. (Terre Jędrzejewski, W., Niedzialkowska, M., Nowak, S., Jędrzejewska, B., 2004. Habitat variables Vie) 52, 357–368. associated with wolf (Canis lupus) distribution and abundance in northern Poland. Randi, E., 2008. Detecting hybridization between wild species and their domesticated Divers. Distrib. 10, 225–233. relatives. Mol. Ecol. 17, 285–293. Kaartinen, S., Luoto, M., Kojola, I., 2009. Carnivore–livestock conflicts: determinants of Randi, E., 2011. Genetics and conservation of wolves Canis lupus in Europe. Mammal Rev. wolf (Canis lupus) depredation on sheep farms in Finland. Biodivers. Conserv. 18, 41, 99–111. 3503–3517. Randi, E., Lucchini, V., Christensen, M.F., Mucci, N., Funk, S.M., Dolf, G., Loeschcke, V., Kernohan, B.J., Gitzen, R.A., Millspaugh, J.J., 2001. Analysis of animal space use and 2000. Mitochondrial DNA variability in Italian and East European wolves: detect- movments. In: Millspaugh, J.J., Marzluff, J.M. (Eds.), Radio Tracking and Animal ing the consequences of small population size and hybridization. Conserv. Biol. Populations. Academic Press, San Diego, pp. 125–166. 14, 464–473. Kovařík, P., Kutal, M., Machar, I., 2014. Sheep and wolves: is the occurrence of large Sand, H., Wikenros, C., Wabakken, P., Liberg, O., 2006. Effects of hunting on group size predators a limiting factor for sheep grazing in the Czech Carpathians? J. Nat. Conserv. snow depth and age on the success of wolves hunting moose. Anim. Behav. 72, 22, 479–486. 781–789. Kruuk, H., Parish, T., 1981. Feeding specialization of the European badger Meles meles in Schenone, L., Aristarchi, C., Meriggi, A., 2004. Ecologia del Lupo (Canis lupus)inProvincia Scotland. J. Anim. Ecol. 50, 773–788. di Genova: distribuzione, consistenza, alimentazione e impatto sulla zootecnia. Landry, J.M., Olsson, P., Siegenthaler, A., Jackson, P., Farrell, A., 1999. The use of guard dogs Hystrix It. J. Mammal. 15, 13–30. in the Swiss Alps: a first analysis. KORA, Rep. n. 2. Seaman, D.E., Millspaugh, J.J., Kernohan, B.J., Brundige, G.C., Raedeke, K.J., Gitzen, R.A., Linnell, J.D., Odden, J., Smith, M.E., Aanes, R., Swenson, J.E., 1999. Large carnivores that kill 1999. Effects of sample size on kernel home range estimates. J. Wildl. Manag. 63, livestock: do ‘problem individuals’ really exist? Wildl. Soc. Bull. 27, 698–705. 739–747. Linnell, J.D., Brøseth, H., Solberg, E.J., Brainerd, S.M., 2005. Theoriginsofthesouth- Sundqvist, A.K., Ellegren, H., Olivier, M., Vilà, C., 2001. Y chromosome haplotyping in ern Scandinavian wolf Canis lupus population: potential for natural immigration Scandinavian wolves (Canis lupus) based on microsatellite markers. Mol. Ecol. 10, in relation to dispersal distances, geography and Baltic ice. Wildl. Biol. 11, 1959–1966. 383–391. Teerink, B.J., 1991. Hair of West European Mammals: Atlas and Identification Key. Lovari, S., Sforzi, A., Scala, C., Fico, R., 2007. Mortality parameters of the wolf in Italy: does Cambridge University Press, Cambridge, UK. the wolf keep himself from the door? J. Zool. 272, 117–124. Thurber, J.M., Peterson, R.O., 1993. Effects of population density and pack size on the Manly, B.F.J., Mcdonald, L.L., Thomas, D.L., Mcdonald, T.L., Erickson, W.P., 2002. Resource foraging ecology of gray wolves. J. Mammal. 74, 879–889. Selection by Animals: Statistical Design and Analysis for Field Studies. second ed. Valière, N., Fumagalli, L., Gielly, L., Miquel, C., Lequette, B., Poulle, M.L., Weber, J.M., Kluwer Press, Dordrecht, Boston, London. Arlettaz, R., Taberlet, P., 2003. Long-distance wolf recolonization of France and 168 C. Imbert et al. / Biological Conservation 195 (2016) 156–168
Switzerland inferred from non-invasive genetic sampling over a period of 10 years. Wang, J., Santure, A.W., 2009. Parentage and sibship inference from multilocus genotype Anim. Conserv. 6, 83–92. data under polygamy. Genetics 181, 1579–1594. van Lière, D., Dwyer, C., Jordan, D., Premik-Banič, A., Valenčič, A., Kompan, D., Siard, N., Wielgus, R.B., Peebles, K.A., 2014. Effects of wolf mortality on livestock depredations. PLoS 2013. Farm characteristics in Slovene wolf habitat related to attacks on sheep. Appl. One 9, 1–16. Anim. Behav. Sci. 144, 46–56. Zimen, E., Boitani, L., 1975. Number and distribution of wolves in Italy. Z. Säugetierkd. 40, Vos, J., 2000. Food habits and livestock depredation of two Iberian wolf packs (Canis lupus 102–112. signatus) in the north of Portugal. J. Zool. 251, 457–462. Zlatanova, D., Ahmed, A., Valasseva, A., Genov, P., 2014. Adaptive diet strategy of the wolf Waits, L.P., Luikart, G., Taberlet, P., 2001. Estimating the probability of identity among (Canis lupus L.) in Europe: a review. Acta Zool. Bulg. 66, 439–445. genotypes in natural populations: cautions and guidelines. Mol. Ecol. 10, Zuur, A.F., Ieno, E.N., Elphick, C.S., 2010. A protocol for data exploration to avoid common 249–256. statistical problems. Methods Ecol. Evol. 1, 3–14. Dead or alive? Comparing costs and benefits of lethal and non-lethal human–wildlife conflict mitigation on livestock farms
J. S. MC M ANUS,A.J.DICKMAN,D.GAYNOR,B.H.SMUTS and D . W . M ACDONALD
Abstract Livestock depredation has implications for con- Introduction servation and agronomy; it can be costly for farmers and – can prompt retaliatory killing of carnivores. Lethal control epredation of livestock is a principal cause of human fl 2001 measures are readily available and are reportedly perceived Dwildlife con ict (Sillero-Zubiri & Laurenson, ). to be cheaper, more practical and more effective than non- It incurs high costs for livestock-keepers and provokes lethal methods. However, the costs and efficacy of lethal vs both retaliatory and preventative killing of carnivores, which non-lethal approaches have rarely been compared formally. may threaten their survival locally or globally (Kruuk, 2002 2005 2013 We conducted a 3-year study on 11 South African livestock ; Ray et al., ; Macdonald et al., ). Pastoralists farms, examining costs and benefits of lethal and non-lethal in the Serengeti have reported that the cost of depredation 19 conflict mitigation methods. Farmers used existing lethal amounts to c. % of their annual cash income (Holmern 2007 control in the first year and switched to guardian animals et al., ), and in Bhutan attacks on livestock by (dogs Canis familiaris and alpacas Lama pacos) or livestock carnivores cost farmers over two-thirds of their annual 2006 protection collars for the following 2 years. During the first cash income, on average (Wang & Macdonald, ). In the year the mean cost of livestock protection was USD 3.30 per USA the annual cost of depredation to the livestock industry 40 2008 head of stock and the mean cost of depredation was is USD million (Conner et al., ). Even greater losses 2010 USD 20.11 per head of stock. In the first year of non-lethal are reported in South Africa, where a survey in esti- control the combined implementation and running costs mated that the annual cost of depredation to the livestock 171 2010 were similar to those of lethal control (USD 3.08 per head). industry is USD million (van Niekerk, ), although a 2007 22 However, the mean cost of depredation decreased by 69.3%, census estimated the cost to be USD million 2010 to USD 6.52 per head. In the second year of non-lethal (Statistics South Africa, ). The disparity between these control the running costs (USD 0.43 per head) were sig- two estimates raises uncertainty as to their accuracy but nificantly lower than in previous years and depredation both reveal a perception that losses to carnivores are high. fi costs decreased further, to USD 5.49 per head. Our results Ideally tools for reducing depredation should bene t suggest that non-lethal methods of human–wildlife conflict both farmers and wildlife conservation. Desirable features ffi mitigation can reduce depredation and can be economically of interventions include persistent e cacy, minimal un- advantageous compared to lethal methods of predator intended environmental consequences, selectivity towards control. problematic individuals, lower cost than that of the de- predation prevented, and social acceptability. Traditionally Keywords Carnivore conservation, conflict mitigation, farmers have attempted to prevent depredation, or retaliate, human–wildlife conflict, lethal control, livestock de- by killing predators (Hone, 1994; Macdonald et al., 2010), predation, non-lethal mitigation techniques, profit/loss ratio often with negative effects on carnivore populations (Sillero- Zubiri & Laurenson, 2001; Woodroffe et al., 2005; Loveridge This paper contains supplementary material that can be et al., 2010). In South Africa, encouraged by the government, found online at http://journals.cambridge.org farmers have employed lethal control of predators, using methods such as gin-traps (leg-hold traps), gun-traps, poison and hunting, with and without hounds, to eradicate carnivores and other problem animals (Daly et al., 2006). As J.S. MCMANUS* (Corresponding author) Centre for African Ecology, School of recently as the 1990s formal bounty systems were in place for Animal, Plant and Environmental Sciences, University of Witwatersrand, Private Bag 3, Johannesburg 2050, South Africa most of the terrestrial mammal species that were perceived E-mail [email protected] to cause conflicts with commercial agriculture, and lethal
A.J. DICKMAN and D.W. MACDONALD WildCRU, Department of Zoology, control is still common on livestock farms (Daly et al., University of Oxford, UK 2006). Despite these measures depredation remains a D. GAYNOR Mammal Research Institute, University of Pretoria, South Africa problem in the livestock farming sector, with indications 2008 B.H. SMUTS Landmark Foundation Trust, Riversdale, South Africa that losses are increasing (Avenant & du Plessis, ). *Also at: Landmark Foundation Trust, PO Box 22, Riversdale 6677, South Africa Lethal control is often considered the cheapest and most effective method of reducing depredation (Conover, 2001; Received 5 June 2013. Revision requested 8 October 2013. Accepted 27 November 2013. Mitchell et al., 2004) but it is not without problems: it may
© 2014 Fauna & Flora International, Oryx, Page 1 of 9 doi:10.1017/S0030605313001610 http://journals.cambridge.org Downloaded: 01 May 2014 IP address: 197.79.56.208 2 J. S. McManus et al.
miss problem individuals and it often fails to eradicate Carnivores typically kill their prey by a fatal bite to the neck, depredation (Avenant & du Plessis, 2008) and involves and these collars protect the vulnerable neck area, increase ongoing commitment and expense (Conover, 2001; Mitchell the effort needed by predators to kill livestock, and reduce et al., 2004). It is commonly unselective and there is little the likelihood of a fatal bite. evidence of cost-effective diminution of livestock losses Such methods have the potential to be more selective (Treves & Naughton-Treves, 2005; Berger, 2006), as pre- than lethal predator control, targeting only those animals dators learn to avoid control efforts (Brand et al., 1995; attempting to kill livestock. There is evidence that non- Knowlton et al., 1999). Methods such as leg-hold traps, lethal interventions can reduce depredation (Breitenmoser snaring and poisoning are largely indiscriminate and often et al., 2005), with the added benefits of favourable public kill non-target species (Rochlitz et al., 2010); in South Africa, perception, improved animal welfare and reduced non- this includes threatened species such as Cape vultures Gyps target casualties (Treves & Naughton-Treves, 2005). Non- coprotheres (Bamford et al., 2007). Unintended outcomes of lethal methods may be more compatible with conservation removing territorial predators can include an influx of objectives and less likely to trigger perturbation effects, replacement individuals, potentially increasing the local including counter-productive ecological cascades such as predator population and the risk of depredation (Crooks & mesopredator release (Beasom, 1974; Crooks & Soulé, 1999). Soulé, 1999; Knowlton et al., 1999), through what are known However, some controls can have negative consequences: generically as perturbation effects (Tuyttens & Macdonald, from the early 1900s to the 1960s most farms in South Africa 2000). Furthermore, lethal control has led to the extermi- were fenced to prevent depredation (Beinart, 2008) but nation of populations of large carnivores (Kruuk, 2002), fencing large areas may restrict the movement of wildlife resulting in debates amongst conservationists, farmers and (Knowlton et al., 1999). Livestock guardian dogs may attack the general public (Treves & Karanth, 2003; Graham et al., wildlife if not properly managed (Green et al., 1984). 2005). Furthermore, although comprehensive audits are few, non- The inadequacies of lethal control methods have focused lethal mitigation techniques are sometimes considered attention on possible non-lethal interventions. One advan- more expensive (Mitchell et al., 2004) and less long-lasting tage of non-lethal control for territorial species is that it does than lethal predator control (Shivik, 2006). There is a dearth not cause social perturbation in the way that lethal control of rigorous accounting of the full life-cycle costs and benefits can do; instead, the target individual is allowed to remain in of alternative interventions. Here we assess the efficacy and its territory and although its behaviour may be altered (e.g. economics of lethal control of carnivores compared to three in the case of learned food aversions) other ecological non-lethal mitigation techniques over a 3-year period. relationships remain intact, including exclusion of poten- fi 1999 tially invading conspeci cs (Reynolds, ). Possible non- Study area lethal interventions include corralling livestock during periods of vulnerability (Schiess-Meier et al., 2007), in- Our study took place on 11 commercial livestock farms at stalling predator-proof fencing around small vulnerable altitudes of 500–2,000 m in the Eastern Cape Province of areas (Breitenmoser et al., 2005), using shepherds (Shivik, South Africa (Fig. 1). Mean farm size was 4,291 ha (1,500– 2006), installing fladry (Davidson-Nelson & Gehring, 2010), 10,000 ha) and the farms covered a total of 47,200 ha and translocating species (Bradley et al., 2005) and using con- received annual rainfall of 230–480 mm. During the period ditioned taste aversion (Cox et al., 2004), other learned food of the study no unusual climatic conditions were experi- aversions (Macdonald & Baker, 2004; Baker et al., 2008) enced. There was extensive grazing on all farms apart or odour (Atkinson & Macdonald, 1994), chemical, visual or from Farm 2, where high-intensity, short-duration planned acoustic repellents (Mason et al., 2001). Guardian animals, grazing rotation was employed (Savory, 1983). Farms particularly livestock guardian dogs, are another popular were subject to varying degrees of depredation and differ- method and have been found to decrease depredation by ent environmental conditions, and represented at least 10–100% on ranches in the USA (Linhart et al., 1979; one of four major biomes: (1) Albany thicket, with rela- Coppinger et al., 1988; Andelt & Hopper, 2000; Gehring tively dense, woody vegetation of mean height c. 2–3 m et al., 2010). They have also proved effective in southern (Mucina & Rutherford, 2005), dominated by Portulacaria Africa: in Namibia, 73% of farmers who used guardian afra (spekboom) and Rhus sp.; (2) Afrotemperate forest dogs reported a significant decline in livestock depredation (Mucina & Rutherford, 2005), with yellowwood Afrocarpus (Marker et al., 2005, 2010). Other guardian animals that falcatus and forest elder Nuxia floribunda in mountainous behave aggressively towards stock-predators can also be gorges; (3) Sandstone fynbos, which was prolific on higher- effective, such as donkeys Equus africanus asinus, alpacas altitude farms and commonly included Protea and Erica; Lama pacos and llamas Lama glama (Conover, 2001). and (4) Nama-Karoo, which is characterized by low A promising but largely untested technique is the use of sweet thorn Acacia karroo in annual river beds, shrubs protective collars made of a strong epoxy–metal mesh. intermixed with grasses, and succulent plants. All farms had
© 2014 Fauna & Flora International, Oryx, 1–9
http://journals.cambridge.org Downloaded: 01 May 2014 IP address: 197.79.56.208 Dead or alive? 3
E E E E
Northern Cape Eastern Cape Graaf Reinet (n = 1) S NorthernWestern Cape Cape
(a) South Africa
Eastern Cape
S Jansenville (n = 2)
Zimbabwe Baviaanskloof (n = 7) Botswana Glenconner (n = 1) Namibia Western Cape
Western Cape FIG. 1 The distribution and number of South Africa S trial farms in the Eastern Cape Province, Legend South Africa. The black shaded area in Eastern Cape Farms (a) indicates the location of South Africa, Formally Protected Areas 0 25 50 100 150 km and (b) indicates the location of the (b) Eastern Cape Province Eastern Cape Province. black-backed jackal Canis mesomelas and caracal Caracal if the death was caused by a predator and, if so, by which caracal present and Farms 5–11 also had leopard Panthera species. If there was doubt, photographs were taken and pardus present. conclusions were made by external experts. Because of the size of the farms, sometimes carcasses were not discovered Methods until it was impossible to determine the cause of death. These were excluded from analyses and cause of death was We conducted interviews with farmers from the 11 farms recorded as unknown. in September 2007, 2008 and 2009. The semi-structured Initial data were collected during August 2006–August questionnaire focused on farm description, quantity of 2007 (the lethal-control year) and all farms converted to livestock, number of losses and their perceived causes, de- non-lethal control in September 2007. Follow-up surveys predation control methods, expenditure on control meth- were conducted at the end of September 2008 to collect data ods, and willingness to adopt non-lethal control methods. on the first non-lethal year, and at the end of September The same questionnaire was administered in person by the 2009 for the second non-lethal year. Therefore, the dataset same researcher each year. for the first non-lethal year includes the initial conversion During the lethal-control year all farms except 1, 4 and from lethal to non-lethal methods. Three farms received 10 used gin-traps and hunting as their control methods. livestock guardian dogs (two received one dog each and the Farm 1 used gun-traps in addition to these methods, Farm 4 other received five), one farm received seven alpacas and used only gin-traps, and Farm 10 used only hunting. During the remaining seven farms received ‘Dead-Stop’ livestock the non-lethal control years implementation and running protection collars (Klaas Louw, Cape Town, South Africa) costs (e.g. veterinary costs and food for livestock guardian for all stock. animals) were sponsored for nine farms for the duration of Farmers received a one-off payment in the first year of the study, after which each farmer took responsibility for implementation to cover the cost of purchasing guardian any costs. Two farms (2 and 4) chose to pay the imple- animals or collars (USD 553 per dog, USD 860 per alpaca mentation and running costs of their preferred control and USD 3.50 per collar). Ongoing maintenance costs for methods for the duration of the study and thereafter. The guardian animals, such as feeding and veterinary care, recruitment of farmers was facilitated by sponsorship of averaged USD 432 per dog and USD 98 per alpaca per year. the controls. There were no running costs for collars in the first year but The allocation of the various non-lethal controls was in the second year there was an additional replacement cost based on the farmers’ willingness to work with livestock of USD 0.35 per collar for wear and tear (10% replacement). guardian animals, and on local conditions. To ensure Maintenance costs for guardian animals remained the same accurate identification of causes of livestock losses, farmers in the second year. If the number of livestock increased attended training workshops and received detailed identifi- between the first and second years of non-lethal control, cation kits and descriptive manuals on kill identification costs for additional collars (one per additional stock animal) (Smuts, 2008). When livestock was depredated, trained were included in year two. The cost of lethal control varied conservation officials and/or one of the researchers, and according to the different methods used by farmers. The the farmer, undertook carcass inspections to determine cost of tools such as gin-traps and gun-traps was calculated
© 2014 Fauna & Flora International, Oryx, 1–9 http://journals.cambridge.org Downloaded: 01 May 2014 IP address: 197.79.56.208 4 J. S. McManus et al.
as the cost of one labourer at minimum wage (USD 8.20 per The mean cost of implementing non-lethal techniques day) because legally the devices must be checked once per was USD 2.91 per head of stock (Supplementary Table S1). day and most farmers assigned one worker to check and set During the first year of non-lethal control the mean running these devices. The cost of hunting was calculated based on cost was USD 0.17 per head of stock (Supplementary daily rates charged by professional vermin hunters (USD 79) Table S1), or USD 336.76 per farm (range USD 0–2,160; there and the number of days these hunters were employed were no running costs during the first year for farms using (2–12 days per year). For each individual of a target species collars; Supplementary Table S1). The mean combined (jackal or caracal) shot by the hunter, an additional USD 122 implementation and running cost during this year was USD was charged. To standardize depredation costs, the cost of 3.08 per head, similar to the running costs of lethal control one depredated animal was calculated at USD 147.42, the (Z 5 −255,P5 0.799). During the same year depredation mean price for a weaned lamb. We use November 2011 prices was significantly lower than when using lethal control and the exchange rate at that time of ZAR 8.14 5 USD 1. (Z 5 2.93,P5 0.003). The mean decline in depredation was Total costs during the lethal-control year were calculated 69.3%, with depredation accounting for 4.4% of stock (range as the sum of running costs and depredation costs, in the 0.1–15.0%; Supplementary Table S2) and costing farmers a first year of non-lethal control as the sum of implemen- mean of USD 6.52 per head of stock (Supplementary tation, running costs and depredation, and in the second Table S1). non-lethal control year as running costs plus depredation. Therefore, the mean total cost per head during the As lethal control had been used prior to the study, im- first year of non-lethal control was USD 9.60 (range USD plementation costs for equipment such as gin-traps, 1.49–28.82; Supplementary Table S1), significantly less gun-traps and poisons were not accounted, and therefore (59.0%) than the cost when using lethal control the overall cost of lethal control may be under-estimated. (Z 5 −2.85,P5 0.004). The cost decreased on 10 of the However, the running costs were considered a close 11 farms (range 41.8–89.9%) but on one farm (Farm 10) there representation of overall costs because items such as gin- was an 8.1% increase in costs relative to the lethal-control traps lasted several years and hunting was calculated as a year (Supplementary Table S1). There was no significant service rather than permanent equipment. Data were not difference in the decline in costs per head between 2 normally distributed so we represented the range of variance farms using alpacas, dogs or collars (χ 5 4.33,df5 2, of the results in the data between the comparative sites. P 5 0.115). However, when comparing our results with other studies The second year of non-lethal control involved no that used mean data we used means in describing the central implementation costs; mean running cost was USD 0.43 tendency. The Wilcoxon signed-rank test for related per head (Supplementary Table S1). This was significantly samples was used to compare different years at the same lower than both the running costs of lethal control 2 sites and the Kruskal–Wallis χ test was used to compare (Z 5 −2.85,P5 0.004) and the combined running and continuous variables between different farms. All analyses implementation costs during the first year of non-lethal were conducted using SPSS v. 16.0 (SPSS, Chicago, USA). control (Z 5 −2.94,P5 0.003), although the costs were significantly higher than the running costs (excluding fi Results implementation costs) of the rst year of non-lethal control (Z 5 −2.31,P5 0.021; Supplementary Table S1). 0 1–14 2 Costs of lethal predator control and non-lethal Depredation, which accounted for . . % of the herd 3 7 2 conflict mitigation (mean . %: Supplementary Table S ), at a mean cost of USD 5.49 per head (Supplementary Table S1), declined During the year of lethal control the cost of control by 72.7% compared to the lethal-control year, which is a measures was USD 3.30 per head of stock and the mean significant difference (Z 5 −2.93,P5 0.003; Supplementary cost of depredation was USD 20.11 per head of stock Table S2). Therefore, the cost of depredation declined by (Supplementary Table S1). With a mean total cost of USD a mean of 15.8% compared to the first year of non-lethal 23.41 per head of stock (Supplementary Table S1), the cost control (Z 5 −1.79,P5 0.074), although on two farms to each farmer was USD 3,552–69,290 depending on their using collars, depredation levels increased between the first stock holdings (mean USD 29,046). There was no significant and second years of non-lethal control (Supplementary difference in total cost per head between farms that would Table S2). Based on stock holdings, during the second year later receive livestock guardian dogs, alpacas or collars of non-lethal control depredation declined by 73.9% 2 (χ 5 3.81,df5 2,P5 0.149). When implementing lethal compared to the lethal-control year and by 13.3% compared control farmers lost 4.0–45% of their stock (mean 13.6%; to the first year of non-lethal control. Supplementary Table S2) to depredation, which, given their The mean total cost per head of non-lethal control in the stock holdings (Supplementary Table S1), equated to a mean second year was USD 5.92 (range 0.72–21.62; Supplementary cost of USD 25,306 per farm (range USD 3,392–66,340). Table S1); this was significantly lower than the cost
© 2014 Fauna & Flora International, Oryx, 1–9
http://journals.cambridge.org Downloaded: 01 May 2014 IP address: 197.79.56.208 Dead or alive? 5 during the lethal-control year (Z 5 −2.93,P5 0.003). year and nine (82%) experienced slightly lower depredation All farms reported lower total costs than during the (Supplementary Table S2). Compared to the first year of lethal-control year, with a mean saving of 74.6% (range non-lethal control, these changes amounted to a profit : loss 54.1–95.1%; Supplementary Table S1). Overall costs were also ratio of 1.23 : 1 (Supplementary Table S3). This ratio did not significantly lower than during the first year of non-lethal differ significantly according to the non-lethal method 2 control on all farms (Z 5 −2.93,P5 0.003), with a mean implemented (χ 5 0.90,df5 2,P5 0.637). decline of 43.9% (range 25.0–76.7%; Supplementary Table S1). The saving per head in the second year of non-lethal control did not differ significantly between Post-trial follow ups farms using different forms of control, when compared 2 Observations made 13 months after the study finished to the lethal-control year (χ 5 2.04,df5 2,P5 0.360)or 2 revealed that 55% of the farms continued to use non-lethal the first year of non-lethal control (χ 5 2.51,df5 2, control. All farms with livestock guardian animals P 5 0.285). (Farms 1–4) retained them at their own expense, two others (6 and 11) acquired livestock guardian animals in addition Avoided depredation and profit : loss ratios in to existing methods, and one (Farm 10) only retained the different years collars. Just under half the farms (45%) combined both lethal and non-lethal methods after cessation of the trial; Although switching from lethal to non-lethal control Farms 5, 7, 8 and 9 used both hunters and collars and Farm 3 resulted in significant declines in both depredation and used gin-traps, hunting dogs, hunting and collars. total costs, this does not fully reflect the economic savings After 30 months 36% of farms (2, 4, 8 and 10) used only that were made. The non-lethal measures led to consider- non-lethal control, 46%(1, 3, 5, 6 and 11) combined lethal able cost savings through avoided depredation, assuming and non-lethal control, and 18%(7 and 9) used only lethal that depredation would have remained at the same level as control. Depredation was reported to have remained the under lethal control. Implementing non-lethal control saved same by 30% of farms (4, 5 and 8), 30% reported an increase farmers a mean of USD 13.58 per head of stock in avoided in depredation (7, 9 and 11) and 40% reported a decrease depredation (Supplementary Table S3), which equates to a (1, 2, 3 and 10) since the end of the trial. This information saving of USD 20,384 per farmer, based on the mean herd was not available for Farm 6 because livestock farming was size of 1,501 in the first year of non-lethal control. Combined only reinstated 1 month prior to the interview. Six farms with a saving on running costs of USD 0.21 per head of used livestock guardian animals (1–4, 6 and 11) but the dog stock (Supplementary Table S3), the overall saving com- on Farm 1 was shot by a neighbour who feared it would pared to what would have been expected under lethal cause damage to livestock. The farmer did not replace the control was USD 13.79 per head, a mean saving of USD dog but instead placed lambing ewes in fenced camps 20,699 per farmer. Given the total cost of implementation, to avoid losses at vulnerable times. Farms 3 and 4 made no running expenses and depredation during the first year, this changes to management and Farms 6 and 11 stopped gives a mean profit : loss ratio of 2.11 : 1, with all but one farming livestock 19 months after the trial ceased, until farmer showing a profit (Supplementary Table S3). Where April 2013 and December 2012, respectively. When livestock the running cost of lethal control was cheaper than that of farming was re-established Farm 6 re-acquired a livestock non-lethal techniques, the lower-than-expected depredation guardian dog and Farm 11 used shepherds and electrified still resulted in a profit (Supplementary Table S3). There was lambing camps. Farms 7 and 9 used only lethal controls. no significant difference in the profit : loss ratio between 2 Farm 9 reported that it was easier to implement because different forms of non-lethal control (χ 5 1.82,df5 2, it was managed by a neighbour; Farm 7 undertook call- P 5 0.403). and-shoot hunting over several farms to reduce predator In the second year of non-lethal control, farmers saved a numbers and avoid losses. Farm 8 used three livestock mean of USD 17.41 per head of stock (range USD 3.29–47.67; guardian dogs and Farm 10 used shepherds in mountainous Supplementary Table S3) compared to what would have areas. All farms except Farm 7 remained willing to pay been expected under lethal control. All farms had a positive for non-lethal controls; Farm 7 indicated that payment profit : loss ratio compared to lethal control, saving a mean to use non-lethal control would increase the likelihood of of USD 5.36 for every USD 1 spent (range USD 1.16–18.11; its use. Supplementary Table S3). As in the first year of non-lethal control, there was no difference in the mean profit : loss ratio between farms using different non-lethal methods 2 Discussion (χ 5 2.04,df5 2,P5 0.360). Comparing the 2 years of non-lethal control, two farms Large carnivores are often highly valued at a global scale but (18%) experienced higher levels of depredation in the second have a low or negative economic value at a local scale
© 2014 Fauna & Flora International, Oryx, 1–9 http://journals.cambridge.org Downloaded: 01 May 2014 IP address: 197.79.56.208 6 J. S. McManus et al.
(Dickman et al., 2011). To address this, local revenue from have occurred without our experimental intervention, or carnivore presence should outweigh the costs of coexistence. that such a reduction occurred on comparable farms that This can be achieved by generating more local revenue, continued to use lethal control. Farms neighbouring the for instance through tourism or payments for presence study farms may have increased the intensity of their (Sillero-Zubiri et al., 2004; Dickman et al., 2011), and also by predator control but we have no evidence for this and it reducing the costs incurred locally as a result of carnivore seems unlikely that this would have happened across all the presence. Our findings suggest that non-lethal mitigation disparate locations. Furthermore, seven of the experimental can effectively reduce depredation and the economic costs farms (Farms 5–11) were adjacent to protected areas, where of carnivores in the vicinity of livestock farming. Farmers there would have been no scope, legally, for control of saved 55.1 and 74.6% during the first and second years of predators; the remaining four had neighbours that practised non-lethal control, respectively, compared to expected lethal control. Some other confounding factor, such as losses during lethal control. Even where lethal controls infectious disease, could have reduced predator populations were cheaper to implement than non-lethal methods, the during the second and third years of our trial on all sampled lower-than-expected depredation resulted in savings in both farms but there was no evidence for this and it is unlikely to years when non-lethal controls were used. There was a mean have occurred at all of the different sites. Farmers may have saving of USD 13.79 per head of stock in the first year of exaggerated reported losses during the first year of the non-lethal control and USD 17.41 per head in the second, survey to demonstrate their need for help, thereby distorting compared to what would be expected when using lethal our findings, but this is unlikely to have occurred in every control only. Overall, farmers saved a mean of . USD case. Although other studies have identified a positive 20,000 during the first year of switching to non-lethal correlation between carnivore absence and human presence measures, which was equivalent to the value of 138 livestock. (Ogada et al., 2003; Bunnefeld et al., 2006) there is no Initiating and operating non-lethal control during the evidence that local human activity was substantially higher first year was cheaper than continuing lethal control on in the years when non-lethal control was implemented. the majority of study farms, and depredation rates were Another possibility is that experimental cessation of lethal invariably lower. In short, non-lethal measures were controls somehow diminished losses of non-target species, cheaper than lethal control on 91% of the farms in and consequently natural prey numbers increased more the first year of implementation. On the one farm where than the target predators, relieving the pressure on domestic the implementation costs were lower for lethal control, stock. Although none of these potential explanations only low-intensity control was employed (Supplementary appears to be likely, we suggest that future studies are run Table S1). In the second year, depredation remained low, with control sites and non-lethal trials concurrently running costs were minimal and all farms reported lower for longer periods, to determine if and when predators costs per head than under lethal control. The economic either adapt to non-lethal measures (Brand et al., 1995)or case for non-lethal approaches is further strengthened if repopulate to a level at which the control measures become the avoided losses from depredation under lethal control ineffective (Gese, 2005), and whether depredation and costs are considered. This made non-lethal predator control twice of control remain low in the long term. Understanding the as lucrative during the first year and . 5 times so during effect of lethal controls on non-target species population the second year. densities could help to determine whether changes occur Large-scale, intensive and expensive lethal-control in their availability and frequency and whether this could experiments have resulted in a 51–68% reduction in affect depredation on livestock. depredation rate (Guthery & Beasom, 1978;O’Gara et al., Although, under the particular circumstances of our 1983; Wagner & Conover, 1999; Greentree et al., 2000). This trial, non-lethal methods yielded significant cost-savings, benefit is similar to, or less than, the 69.3 and 73.9% using lethal control is not purely an economic decision. reduction we found during the first and second years of Hunting of carnivores is often culturally and socially implementing non-lethal measures. Given the higher cost embedded and may provide intangible benefits such as of lethal control, this suggests that non-lethal measures are social prestige and enjoyment (Hazzah et al., 2009; Marchini a more economical option. & Macdonald, 2012). It is likely that many farmers will want We are mindful that the design of our study lacked a to continue some form of carnivore hunting but if non- formal control, as there was no sample of farms in the lethal methods continue to be effective it will be clear that second and third years on which lethal control continued this is driven more by cultural norms and satisfaction for comparison. Nonetheless, the reductions in cost and than economics. Furthermore, the adoption of non-lethal depredation were similar across all farms (irrespective methods will depend on the local context; for example in of locations, biomes and environmental conditions). We many areas livestock-keepers may not have the means or the do not have grounds to believe that the reduction in inclination to invest time, care and resources in livestock depredation that occurred during non-lethal control would guardian dogs or exotic animals such as alpacas.
© 2014 Fauna & Flora International, Oryx, 1–9
http://journals.cambridge.org Downloaded: 01 May 2014 IP address: 197.79.56.208 Dead or alive? 7
Determining the most appropriate methods for the local Acknowledgements socio-economic and cultural environment is a vital step in encouraging farmers to adopt novel forms of non-lethal We are grateful to the private landowners who willingly control. According to our interviews all 11 farmers were participated in this trial of non-lethal predator control and willing to pay for non-lethal control if it effectively reduced agreed to forgo the use of lethal controls for the duration of losses. However, 9 of the 11 farms were provided with such the study, many of whom have continued to use non-lethal controls free of charge, which probably accelerated the controls following the trial. We thank Sandra Baker, Paul rate of uptake, and two farmers bought and implemented Johnson, Gus Mills, Jan Kamler and Alan Lee for insightful the controls without financial support. Therefore, although comments, and the ABAX Foundation (previously the ’ many farmers may be willing to implement non-lethal Polaris Foundation), Pick n Pay, Woolworths, the Henry controls they may not do so without incentives or support and Iris Englund Foundation, the National Lotteries of some kind. However, attitudes towards predators are Distribution Trust Fund, Arne Hanson, the Mones rarely based on economics alone but are influenced by Michaels Trust and Royal Canin for sponsoring the tools – a variety of personal factors, including beliefs and values, to undertake this research. A Wits Carnegie fellowship fi education, upbringing, tradition and culture (Zimmermann provided support to JM during the nal preparation of the et al., 2005). article, AD is a Kaplan Senior Research Fellow at Pembroke Our observations indicate that after non-lethal controls College, Oxford, and DWM warmly acknowledges the – ’ are introduced, in most cases (82%) their use is continued support of the Recanati Kaplan Foundation, the Peoples or alternative non-lethal methods are tried, either in Trust for Endangered Species and the Swift family. isolation or alongside lethal controls. Depredation increased on the two farms where only lethal controls were used and References decreased on 50% of farms where only non-lethal methods were implemented. On the other 50% there was no change ANDELT,W.F.&HOPPER, S.N. (2000) Livestock guard dogs reduce in the level of depredation. Where lethal and non-lethal predation on domestic sheep in Colorado. Journal of Range Management, 53, 259–267. controls were combined, losses to depredation decreased 1994 50 25 ATKINSON, R.P.D. & MACDONALD, D.W. ( ) Can on % of farms, remained the same on % of farms and repellents function as a nonlethal means of controlling moles increased on the remaining 25%. Given that depredation (Talpa-Europaea)? Journal of Applied Ecology, 31, 731–736. increased when using lethal controls, it seems that the use AVENANT, N.L. & DU PLESSIS, J.J. (2008) Sustainable small stock of such controls is influenced by the attitudes of farmers farming and ecosystem conservation in Southern Africa: a role 72 258–263 and their neighbours as much as by any realized economic for small mammals? Mammalia, , . BAKER, S.E., ELLWOOD, S.A., SLATER, D., WATKINS, R.W. & advantages. MACDONALD, D.W. (2008) Learned food aversion plus odour Further and long-term controlled trials are needed cue protects crop from mammalian wildlife foraging. The Journal to investigate whether the benefits we observed as a result of Wildlife Management, 72, 785–791. of non-lethal controls are sustainable. Such trials could BAMFORD, A.J., DIEKMANN, M., MONADJEM,A.&MENDELSOHN,J. 2007 also evaluate predator habituation to non-lethal techniques ( ) Ranging behaviour of Cape Vultures Gyps coprotheres from ff an endangered population in Namibia. Bird Conservation and the e ectiveness of methods such as using alpacas to International, 17, 331–339. guard against larger predators such as leopards. Our BEASOM,S.L.(1974) Selectivity of predator control techniques results suggest that non-lethal forms of livestock protection, in South Texas. The Journal of Wildlife Management, 38, whether livestock guardian animals or barriers such as 837–844. 2008 collars, can efficiently and cost-effectively reduce de- BEINART,W.( ) The Rise of Conservation in South Africa: Settlers, Livestock, and the Environment 1770–1950. Oxford University Press, predation on domestic stock. These methods reduced the Oxford, UK. economic cost of livestock depredation by carnivores for at BERGER, K.M. (2006) Carnivore–livestock conflicts: effects of least 2 years, which is important for improving the local subsidized predator control and economic correlates on the sheep cost : benefit ratio of carnivore presence. They may also industry. Conservation Biology, 20, 751–761. benefit conservation by reducing the motivation for BRADLEY, E.H., PLETSCHER, D.H., BANGS, E.E., KUNKEL, K.E., SMITH, D.W., MACK, C.M. et al. (2005) Evaluating wolf retaliatory or pre-emptive killing of carnivores, and by fl ff translocation as a nonlethal method to reduce livestock con icts reducing the e ects of control on non-target species. The in the northwestern United States. Conservation Biology, 19, use of non-lethal conflict mitigation approaches may also be 1498–1508. useful in reducing edge effects (Woodroffe & Ginsberg, BRAND, D.J., FAIRALL,N.&SCOTT, W.M. (1995) The influence 1998) on species and provide safety buffer zones adjacent of regular removal of black-backed jackals on the efficiency of to protected areas or along important wildlife corridors. coyote getters. South African Journal of Wildlife Research, 25, 44–48. Since this study was completed there has been a strong BREITENMOSER, C., ANGST, U., LANDRY, C., BREITENMOSER- uptake in the use of various livestock protective collars, WÜRSTEN, C., LINNELL, J.D.C. & WEBER, J.M. (2005) including in Iran. Non-lethal techniques for reducing depredation. In People
© 2014 Fauna & Flora International, Oryx, 1–9 http://journals.cambridge.org Downloaded: 01 May 2014 IP address: 197.79.56.208 8 J. S. McManus et al.
and Wildlife: Conflict or Coexistence? (eds R. Woodroffe, KNOWLTON, F.F., GESE, E.M. & JAEGER, M.M. (1999) Coyote S. Thirgood & A. Rabinowitz), pp. 49–71. Cambridge University depredation control: an interface between biology and management. Press, Cambridge, UK. Journal of Range Management, 52, 398–412. BUNNEFELD, N., LINNELL, J.D.C., ODDEN, J., VAN DUIJN, M.A.J., KRUUK,H.(2002) Hunter and Hunted: Relationships between &ANDERSEN,R.(2006) Risk taking by Eurasian lynx (Lynx lynx) Carnivores and People. Cambridge University Press, in a human-dominated landscape: effects of sex and reproductive Cambridge, UK. status. Journal of Zoology, 270, 31–39. LINHART, S.B., STERNER, R.T., CARRIGAN, T.C. & HENNE, D.R. CONNER, M.M., EBINGER, M.R., KNOWLTON,F.F.(2008) Evaluating (1979) Komondor guard dogs reduce sheep losses to coyotes: coyote management strategies using a spatially explicit, individual- a preliminary evaluation. Journal of Range Management, based, socially structured population model. Ecological Modelling, 32, 238–241. 219, 234–247. LOVERIDGE, A.J., WANG, S.W., FRANK, L.G. & SEIDENSTICKER,J. CONOVER,M.(2001)Effect of hunting and trapping on wildlife (2010) People and wild felids: conservation of cats and damage. Wildlife Society Bulletin, 29, 521–532. management of conflicts. In Biology and Conservation of Wild COPPINGER, R., COPPINGER, L., LANGELOH, G., GETTLER,L.& Felids (eds D.W. Macdonald & A.J. Loveridge), pp. 161–195. Oxford LORENZ,J.(1988) A decade of use of livestock guarding dogs. University Press, Oxford, UK. Proceedings of the Thirteenth Vertebrate Pest Conference, 13, MACDONALD, D.W. & BAKER, S.E. (2004) Non-lethal control of fox 209–214. predation: the potential of generalised aversion. Animal Welfare, COX, R., BAKER, S.E., MACDONALD, D.W. & BERDOY,M.(2004) 13, 77–85. Protecting egg prey from carrion crows: the potential of aversive MACDONALD, D.W., BOITANI, L., DINERSTEIN, E., FRITZ,H. conditioning. Applied Animal Behaviour Science, 87, 325–342. &WRANGHAM,R.(2013) Conserving large mammals: are CROOKS, K.R. & SOULÉ, M.E. (1999) Mesopredator release and they a special case? In Key Topics in Conservation Biology 2 avifaunal extinctions in a fragmented system. Nature, 400, 563–566. (eds D.W. Macdonald & K.J. Willis), pp. 277–312. Wiley-Blackwell, DALY, B., DAVIES-MOSTERT, H., DAVIES-MOSTERT, W., EVANS, S., Oxford, UK. FRIEDMANN, Y., KING, N. et al. (eds) (2006) Proceedings of a MACDONALD, D.W., LOVERIDGE, A.J. & RABINOWITZ,A.(2010) Workshop on Holistic Management of Human–Wildlife Conflict in Felid futures: crossing disciplines, borders and generations. the Agricultural Sector of South Africa. Endangered Wildlife Trust, In Biology and Conservation of Wild Felids (eds D.W. Macdonald Johannesburg, South Africa. & A.J. Loveridge), pp. 599–649. Oxford University Press, DAVIDSON-NELSON, S.J. & GEHRING, T.M. (2010) Testing fladry as a Oxford, UK. nonlethal management tool for wolves and coyotes in Michigan. MARCHINI,S.&MACDONALD, D.W. (2012) Predicting ranchers’ Human–Wildlife Interactions, 4, 87–94. intention to kill jaguars: case studies in Amazonia and Pantanal. DICKMAN, A.J., MACDONALD, E.A. & MACDONALD, D.W. (2011) Biological Conservation, 147, 213–221. A review of financial instruments to pay for predator conservation MARKER, L.L., DICKMAN, A.J. & MACDONALD, D.W. (2005) Perceived and encourage human–carnivore coexistence. Proceedings of the effectiveness of livestock-guarding dogs placed on Namibian farms. National Academy of Sciences of the United States of America, Rangeland Ecology & Management, 58, 329–336. 108, 13937–13944. MARKER, L., DICKMAN, A., MILLS,G.&MACDONALD, D.W. GEHRING, T.M., VERCAUTEREN, K.C. & LANDRY, J.M. (2010) (2010) Cheetahs and ranchers in Namibia: a case study. In Livestock protection dogs in the 21st century: is an ancient Biology and Conservation of Wild Felids (eds D.W. Macdonald tool relevant to modern conservation challenges? BioScience, & A.J. Loveridge), pp. 353–372. Oxford University Press, 60, 299–308. Oxford, UK. GESE,E.(2005) Demographic and spatial responses of coyotes MASON, J.R., SHIVIK, J.A. & FALL, M.W. (2001) Chemical repellents to changes in food and exploitation. Proceedings of the Wildlife and other aversive strategies in predation management. Endangered Damage Management Conference, 11, 271–285. Species Update, 18, 175–181. GRAHAM, K., BECKERMAN, A.P. & THIRGOOD,S.(2005) MITCHELL, B.R., JAEGER, M.M. & BARRETT, R.H. (2004) Coyote Human–predator–prey conflicts: ecological correlates, prey depredation management: current methods and research needs. losses and patterns of management. Biological Conservation, Wildlife Society Bulletin, 32, 1209–1218. 122, 159–171. MUCINA,L.&RUTHERFORD, M.C. (eds) (2005) The Vegetation GREEN, J.S., WOODRUFF, R.A. & TUELLER,T.T.(1984) Livestock- of South Africa, Lesotho and Swaziland. South African National guarding dogs for predator control: costs, benefits, and practicality. Biodiversity Institute, Pretoria, South Africa. Wildlife Society Bulletin, 12, 44–50. OGADA, M.O., WOODROFFE, R., OGUGE, N.O. & FRANK, L.G. (2003) GREENTREE, C., SAUNDERS, G., MCLEOD,L.&HONE,J.(2000) Lamb Limiting depredation by African carnivores: the role of livestock predation and fox control in south-eastern Australia. Journal of husbandry. Conservation Biology, 17, 1521–1530. Applied Ecology, 37, 935–943. O’GARA, B.W., BRAWLEY, K.C., MUNOZ, J.R. & HENNE, D.R. GUTHERY, F.S. & BEASOM, S.L. (1978)Effects of predator control on (1983) Predation on domestic sheep on a western Montana ranch. Angora goat survival in South Texas. Journal of Range Management, Wildlife Society Bulletin, 11, 253–264. 31, 168–173. RAY, J., HUNTER,L.&ZIGOURIS,J.(2005) Setting Conservation HAZZAH, L., BORGERHOFF MULDER,M.&FRANK,L.(2009) Lions and Research Priorities for Larger African Carnivores. Wildlife and warriors: social factors underlying declining African lion Conservation Society, New York, USA. populations and the effect of incentive-based management in REYNOLDS, J.C. (1999) The potential for exploiting conditioned Kenya. Biological Conservation, 142, 2428–2437. taste aversion (CTA) in wildlife management. In Advances in HOLMERN, T., NYAHONGO,J.&ROSKAFT,E.(2007) Livestock loss Vertebrate Pest Management (eds D.P. Cowan & C.J. Feare), caused by predators outside the Serengeti National Park, Tanzania. pp. 267–82. Filander Verlag, Fürth, Germany. Biological Conservation, 135, 518–526. ROCHLITZ, I., PEARCE, G.P. & BROOM, D.M. (2010) The Impact HONE,J.(1994) Analysis of Vertebrate Pest Control. Cambridge of Snares on Animal Welfare. Cambridge University Press, University Press, New York, USA. Cambridge, UK.
© 2014 Fauna & Flora International, Oryx, 1–9
http://journals.cambridge.org Downloaded: 01 May 2014 IP address: 197.79.56.208 Dead or alive? 9
SAVORY,A.(1983) The Savory grazing method or holistic resource W.J. Sutherland), pp. 315–329. Cambridge University Press, management. Rangelands, 5, 155–159. Cambridge, UK. SCHIESS-MEIER, M., RAMSAUER, S., GABANAPELO,T.&KÖNIG,B. VAN NIEKERK, H.N. (2010) The cost of predation on small livestock (2007) Livestock predation—insights from problem animal in South Africa by medium sized predators. MSc thesis. Free State control registers in Botswana. The Journal of Wildlife Management, University, Bloemfontein, South Africa. 71, 1267–1274. WAGNER, K.K. & CONOVER, M.R. (1999)Effect of preventive coyote SHIVIK, J.A. (2006) Tools for the edge: what’s new for conserving hunting on sheep losses to coyote predation. The Journal of Wildlife carnivores. BioScience, 56, 253–259. Management, 63, 606–612. SILLERO-ZUBIRI,C.&LAURENSON, M.K. (2001) Interactions WANG, S.W. & MACDONALD, D.W. (2006) Livestock predation by between carnivores and local communities: conflict or co-existence? carnivores in Jigme Singye Wangchuck National Park, Bhutan. In Carnivore Conservation (eds J.L. Gittleman, S.M. Funk, Biological Conservation, 129, 558–565. D.W. Macdonald & R.K. Wayne), pp. 282–312. Cambridge WOODROFFE,R.&GINSBERG, J.R. (1998) Edge effects and the University Press, Cambridge, UK. extinction of populations inside protected areas. Science, SILLERO-ZUBIRI, C., REYNOLDS,J.&NOVARO, A.J. (2004) 280, 2126–2128. Management and control of wild canids alongside people. WOODROFFE, R., THIRGOOD,S.&RABINOWITZ,A.(2005) The impact In Biology and Conservation of Wild Canids (eds D.W. Macdonald of human–wildlife conflict on natural systems. In People and & C. Sillero-Zubiri), pp. 123–142. Oxford University Press, Wildlife: Conflict or Coexistence? (eds R. Woodroffe, S. Thirgood New York, USA. & A. Rabinowitz), pp. 1–12. Cambridge University Press, SMUTS,B.(2008) Predators on Livestock Farms: A Practical Cambridge, UK. Farmers’ Manual for Non-lethal, Holistic, Ecologically Acceptable ZIMMERMANN, A., WALPOLE,M.J.&LEADER-WILLIAMS,N.(2005) and Ethical Management. Landmark Foundation, Riversdale, Cattle ranchers’ attitudes to conflicts with jaguar Panthera onca South Africa. Http://www.landmarkfoundation.org.za/uploads/5/ in the Pantanal of Brazil. Oryx, 39, 406–412. 9/0/1/5901915/predators_on_livestock_farms_a_practical_manual_ for_non_lethal_holistic_ecologically_acceptable_and_ethical_ management_landmark_foundation.pdf [accessed 6 March 2014]. Biographical sketches STATISTICS SOUTH AFRICA (2010) Census of Commercial Agriculture 2007. Report No. 11-02-01 (2007). Statistics South Africa, Pretoria, J EANNINE M C M ANUS has experience in restoring wildlife habitat South Africa. and addressing the conflict caused by economic losses to the livestock TREVES,A.&KARANTH, K.U. (2003) Human–carnivore conflict and industry as a result of carnivore predation. A MY D ICKMAN has been perspectives on carnivore management worldwide. Conservation researching human–carnivore conflict in Africa for 15 years, with a Biology, 17, 1491–1499. particular focus on testing the efficacy of non-lethal conflict mitigation TREVES,A.&NAUGHTON-TREVES,L.(2005) Evaluating lethal control techniques. D AVID G AYNOR is a specialist in mammal behaviour in the management of human–wildlife conflict. In People and ecology and is also a livestock farmer. B OOL S MUTS has been involved Wildlife: Conflict or Coexistence? (eds R. Woodroffe, S. Thirgood in human–wildlife conflict mitigation efforts since 2004. He develops & A. Rabinowitz), pp. 86–106. Cambridge University Press, projects integrating environmental ethics, viable economics and Cambridge, UK. social responsibility. D AVID M ACDONALD’s research interests lie in TUYTTENS, F.A.M. & MACDONALD, D.W. (2000) Consequences the behavioural ecology of carnivores, especially with relevance of social perturbation for wildlife management and to human–carnivore conflict. He has studied large felids in Africa, conservation. In Behaviour and Conservation (eds L.M. Gosling & Asia and South America.
© 2014 Fauna & Flora International, Oryx, 1–9 http://journals.cambridge.org Downloaded: 01 May 2014 IP address: 197.79.56.208 Project to Reform Public Land Grazing in Northern California Felice Pace, Coordinator 28 Maple Road Klamath, Ca 95548 707-954-6588 [email protected] A report to sponsors, allies and supporters on efforts in 2015 to Reform Public Land Grazing
By Felice Pace, Project Coordinator
Introduction
In 2015 the Project to Reform Public Land Grazing in Northern California completed its sixth year monitoring grazing management on the ground in Northern California’s national forests. As in past years, our objective was to document the impacts of poorly managed grazing on water quality and habitat and to recommend to responsible Forest Service officials and regional water quality regulators what needs to be done to manage grazing responsibly. Responsible grazing would reduce to insignificance or eliminate the adverse impacts of poorly managed grazing.
Four volunteers logged 224 hours (28 person days) monitoring 14 national forest grazing allotments on-the-ground in 2015. That is nearly double the monitoring hours logged in 2014. In 2015 we monitored grazing allotments from the Oregon-California border to the McCloud Flats South of Mount Shasta.
In the north, three volunteers monitored the Elliot and Beaver-Silver grazing allotments north of the Siskiyou Crest on the Rogue-Siskiyou National Forest and the Horse Creek, Dry Lake and East Beaver Grazing Allotments on the Klamath National Forest south of the Crest. Running west from Mount Ashland, the Siskiyou Crest contains a complex of sensitive lands and roadless areas. Grazing degradation there began in the 1800s and continues today.
The three grazing allotments south of the Crest extend to lower elevation lands near the Klamath River. Both the Beaver and Horse Creek Watersheds include miles of streams that have been designated Critical Habit for Coho Salmon.
Farther south, Project volunteers monitored the Big Meadows grazing allotment in the Marble Mountain Wilderness and the Carter Meadows and Mill Creek grazing allotments in the Trinity Alps Wilderness. Those allotments are on the Klamath National Forest; the Mill Creek allotment includes meadows and headwater basins on both the Klamath and Shasta-Trinity National Forests.
On the following page you’ll find a map on which the twelve grazing allotments on the Klamath National Forest which the Project has monitored are depicted and can be located.
(Please go on to the next page)
While monitoring grazing on the Big Meadows Allotment, I was joined by hydrologist Jon Rhodes. An expert on meadow hydrology and grazing impacts to water quality, the Project hired Jon to study and report on the impact of Big Meadows grazing on water quality, meadow hydrology, water supply, riparian vegetation and wetlands. As I write this annual report we await Jon’s Big Meadows report which we will use to “encourage” Forest Service managers to implement needed grazing management reforms. To learn about Jon’s work around the West to reform and challenge irresponsible and damaging public land grazing see his web site at this link.
There was a Project milestone in 2015 when we monitored grazing in the Mount Shasta area of the Shasta Trinity National Forest (STNF) for the first time. Project volunteer and former Forest Service biologist Francis Mangles, who before he retired managed STNF grazing allotments, joined me in monitoring the Bartle Grazing Allotment on the McCloud Ranger District east of Mt. Shasta and the Bear Creek Grazing Allotment on the north and east slopes of Mount Eddy within the Mount Shasta Ranger District.
Retired Forest Service biologist Francis Mangles examines riparian conditions on the Bear Creek Grazing Allotment near Mount Eddy on the Shasta Trinity National Forest
All on-the-ground monitoring by the Project is 100% volunteer, including my time. Thanks to a grant from the Fund for Wild Nature, the Project was able to reimburse volunteers for mileage and expenses incurred while monitoring.
Funding from generous EPIC donors makes it possible for me to also do paid work on public land grazing for EPIC. That work includes commenting on and challenging Forest Service proposals to reauthorize grazing on North Coast and Klamath River Basin national forests.
EPIC is one of three environmental organizations which sponsor The Project to Reform Public Land Grazing in Northern California; Klamath Forest Alliance (KFA) and Montana-based Wilderness Watch are also sponsors. A grant to KFA from the Giles and Elsie Mead Foundation also supports work on grazing. Mead Foundation and EPIC donor funding makes it possible for the Project to use its monitoring findings to challenge and reform public land grazing.
Over the course of the past six grazing seasons, the Project has monitored 16 separate grazing allotments on 3 national forests: 2 allotments on the Shasta-Trinity National Forest, 2 allotments on the Rogue-Siskiyou National Forest and 12 grazing allotments on the Klamath National Forests. We’ve monitored grazing in the Marble Mountain, Russian Peak and Trinity Alps Wilderness Areas, along the Siskiyou Crest west of Mt. Ashland and in several roadless areas. Most allotments have been monitored multiple-times and in multiple years.
Environmental impacts, including damage to water quality, impairment of meadow hydrology and degradation of fish, amphibian and wildlife habitat, are similar on all the grazing allotments the Project has monitored. Those impacts, however, vary in intensity; the meadows where most grazing occurs are trashed on some allotments while on others they are only damaged. Management deficiencies are also consistent across all 16 grazing allotments the Project has monitored. The failure of Forest Service managers to require responsible grazing and modern grazing methods are the root cause of the degradation of water quality, wetlands, riparian areas and wildlife habitat which we have documented. In the following sections I illustrate those degraded conditions and the management deficiencies which cause the degradation of what should be our highest quality waters.
The environmental impact of grazing on Northern California’s national forests includes degradation of water quality, the fragmentation and degradation of wetlands, the devegetation of dry meadows and negative impacts to wildlife and the habitats on which wildlife depend.
Water Quality Degradation
Grazing on national forest land in Northern California degrades water quality in violation of Clean Water Act standards. Water quality degradation related to poorly managed grazing includes excessive sedimentation, fecal bacteria and nutrient pollution and unnaturally high stream water temperature.
Excessive and nuisance sedimentation is the result of cattle weighing up to and over 1,200 pounds trampling springs and streambanks. When cattle are not herded frequently they will spend an inordinate amount of time grazing while standing and walking in springs, in streams and on streambanks.
Below are photos illustrating the streambank and spring trampling which the Project has found consistently on national forest grazing allotments. On many of the 16 grazing allotments we’ve monitored it is hard to find a spring that has not been trampled and streambank trampling is common in preferred areas where poorly herded cattle remain for long periods.
A trampled streambank on East Boulder Creek, Mill Creek Grazing Allotment in the Trinity Alps Wilderness
Trampling and deposition of bovine waste into Mud Springs near the Siskiyou Crest on the Dry Lake Grazing Allotment
Project volunteer Luke Ruediger examines cattle damage to a riparian area. This is part of the Elliot Creek Grazing Allotment on the Rogue-Siskiyou National Forest.
Poorly managed cattle grazing removes shade which raises the temperature of the water
Over the past seven years the Quartz Valley Indian Reservation, a federally recognized tribe, has monitored water quality in streams issuing from national forest grazing allotments and at other stream locations within the Scott River Basin. The Tribe, which has its own certified water quality laboratory, has documented multiple violations of North Coast Water Board and EPA standards for fecal bacteria pollution as well as nutrient pollution in streams below the grazing allotments. A report summarizing “Patterns in Fecal Indicator Bacteria in the Scott River Watershed, 2007-2014” can be accessed and downloaded at this link.
QVIR water quality testing was the main factor motivating me to begin the Project to Reform Public Land Grazing in Northern California six years ago. It started as an effort to document on- the-ground management failures causing the Clean Water Act violations the Tribe found.
The #1 cause of fecal bacteria and nutrient pollution in lakes and streams on and flowing from national forest grazing allotments is direct deposition of bovine waste into streams and onto streambanks. The photo below shows direct deposition of waste into Taylor Lake, a popular swimming lake in the Russian Peak Wilderness just ½ mile from the trailhead.
Cattle manure in Taylor Lake, a popular swimming lake in the Russian Peak Wilderness
Taylor Lake trail is accessible to individuals with disabilities; the lake is a popular place for locals to take youngsters swimming.
When cattle are poorly managed they spend an inordinate amount of time grazing within the riparian areas adjacent to streams. While on the streams cattle also browse riparian vegetation. One typical result is the removal of shade from the stream. Opening the streams to increased solar radiation raises the water’s temperature which impacts all those species which depend on cold water stream habitat including frogs, other amphibians like the Pacific Giant Salamander, trout and salmon in the streams below.
The Pacific Giant Salamander is a top predator in healthy Northern California headwater streams
The impact of grazing in the headwater basins of key national forest salmon and steelhead watersheds has gone largely unassessed and is denied by Forest Service managers. Managers appear to prefer to not know what might require actions that would likely cause conflict with the ranchers who graze on public lands. In light of what is occurring on the Malheur Wildlife Refuge in eastern Oregon as I write this report, one can understand why most Forest Service grazing managers prefer to ignore the environmental impacts of poorly managed grazing. But avoidance of conflict is not a valid justification for managers who refuse to enforce the regulations and require the practices that protect water quality, public lands and wildlife habitat from degradation.
Wetland degradation and fragmentation
As noted above, when they are not adequately herded to rotate grazing among the various pastures on a grazing allotment, cattle find a preferred area and remain there until they are herded to a different area or until all available forage has been consumed. Often those preferred locations are in lower gradient meadow systems in which there are large wetlands dominated by willow species.
Most healthy Northern California willow wetlands found on national forests are from ½ to 2 acres in extent and from 5 to 15 feet in height. They contain continuous dense vegetation from the tops of the willows to the ground rendering them nearly impenetrable to humans.
(Please go on to the next page)
Healthy, ungrazed willow wetlands in the Marble Mountain Wilderness
Poorly herded cattle push into and through the willow wetlands to get to the tender grasses and sedges growing under the willows. The bovines also browse the willows, especially in fall after the first frosts. When that sort of use is allowed to continue for long periods and year after year the willow stands are progressively fragmented and degraded. Bovine trampling dries out the wetlands over time. In the most extreme cases willow wetlands are being slowly converted into grasslands.
This severely fragmented and degraded willow wetland in the Black Meadows pasture of the Big Meadows Grazing Allotment is being slowly converted into a grassland
Large and dense willow wetlands provide breeding habitat for the Willow Flycatcher, a bird species listed as “endangered” under provisions of the California Endangered Species Act. The diminutive flycatcher is also a “management indicator species” for California national forests. That means every national forest in California is supposed to monitor either the bird itself or its breeding habitat. The forest plan for the Klamath National Forest calls for breeding ground surveys to determine if Willow Flycatchers are present in the large willow wetlands which are the birds breeding habitat. But the Forest Service has mostly failed to conduct those surveys and has never conducted surveys on grazing allotments. Nor does the Klamath National Forest monitor to determine the quality and suitability of Willow Flycatcher breeding habitat. It appears that Forest Service managers would prefer not to know whether what is supposed to be a “management indicator species” is breeding successfully and whether the species’ breeding habitat is suitable or degraded.
In response to the FS failure to monitor Willow Flycatcher breeding and habitat, the Project began conducting breeding ground Willow Flycatcher surveys under established protocols three years ago. We’ve monitored what should be suitable breeding habitat both within and outside national forest grazing allotments. In spite of monitoring before the cattle arrive, which is usually in mid-July, we’ve never heard a Willow Flycatcher respond to the territorial bird calls we’ve broadcast near willow wetlands that are grazed. We have received responses, and once positively identified a Willow Flycatcher, when we’ve monitored suitable habitat outside grazed areas.
According to the scientific literature on Willow Flycatcher, the main reason the diminutive bird is C-ESA listed and a management indicator species is its sensitivity to grazing. When cattle are allowed to push into and fragment large willow wetlands, the bird’s nesting habitat (dense vegetation in the interior of the stands within 5 feet of the ground) is destroyed. The Project has confirmed what is stated in the scientific literature: poorly managed grazing destroys Willow Flycatcher nesting habitat, extirpating the species from national forest grazing allotments.
Dry meadow devegetation
In Northern California and throughout the West most dry meadows and grasslands in a natural condition are dominated by native bunchgrass. Ungrazed bunchgrass stands are dense, vibrant and quite beautiful, particularly in springtime which, in the higher mountains where most public land grazing in Northern California occurs, can extends into August.
A healthy bunchgrass stand in an ungrazed portion of the Marble Mountain Wilderness
Unfortunately western bunchgrasses, which co-evolved with widely ranging herds of native ungulates, are particularly sensitive to repeated grazing during a single season. That makes the bunchgrasses vulnerable to cattle because, unlike native ungulates, cattle do not naturally range widely, especially if they are not herded regularly.
Forest Service managers do not require regular herding. As a result, native bunchgrass stands within Northern California national forest grazing allotments are degraded. In many of the preferred areas where unherded cattle congregate for long periods, bunchgrass has been totally extirpated from large areas of dry meadow creating barrens. The loss of dryland bunchgrasses on national forest grazing allotments has made the cattle which graze there more dependent on riparian areas and wetlands and that accelerates degradation of water quality, streambanks and wetlands.
In the preferred areas where unherded cattle congregate for long periods bunchgrasses have been wiped out and barrens have been created.
The Project has monitored a few national forest locations where grazing ended several years back because there were no ranchers who wanted those allotments. We’ve watched bunchgrass stands in dry meadows recover slowly in those locations once grazing ends. Preventing cattle from grazing a dry meadow more than once during a single season might lead to bunchgrass recovery. But that is not the way the national forest grazing allotments the Project has monitored are managed.
Competition with wildlife
Reintroduced and recovering elk herds are expanding their range across Northern California. The expansion has brought elk back to many of the grazing allotments which the Project monitors. Like cattle, elk are primary grazers which also browse on woody vegetation. Unlike cattle, elk do not remain in preferred locations for long periods; except for mothers when they are with young foals, elk typically move constantly within their territories. In those portions of the national forests where elk and cattle range overlaps, they compete for the available forage. Responsible Forest Service managers would reduce the number of cattle allowed to graze when elk reoccupy a grazing allotment because there is only so much forage available in a given year. Failure to adjust what is referred to as cattle “stocking” when elk reoccupy national forest land where livestock grazing occurs results in overgrazing and accelerated degradation of meadows, riparian areas and willow wetlands.
Elk sign near Reeves Ranch Spring along the Siskiyou Crest on June 1, 2015. Horse Creek Allotment cattle had not yet reached this location but elk had already consumed most of the available forage.
The following two photos show two of the classic indicators of overgrazing: trailing and pedestalling. The Project has documented trailing and pedestalling on most of the grazing allotments we’ve monitored.
“Trailing” which refers to the multiple trails which cattle sometimes create is a classic indicator of overgrazing.
“Pedestalling” refers to vegetation perched on a soil pedestal. It is another classic indicator of overgrazing
Forest Service (mis)management
In describing the environmental degradation which Project monitors have documented, I have often referred to “poorly herded” and “unherded” cattle or livestock. The technical term for managing livestock without regular herding to move them out of wetlands and rotate them among the various pastures on a grazing allotment is “passive, season-long grazing.”
“Passive season-long grazing’ is not a valid grazing management method but rather the lack of a method. On every one of the sixteen national forest grazing allotments we’ve monitored, “herding” is mentioned in Allotment Management Plans and in the Annual Operating Instructions issued each year to grazing permit holders. But reality on the ground is that herding is only required when there is a complaint in which case a Forest Service range technician or ranger will ask the permit holding rancher to go out and move the cattle. The ranchers usually, but not always, comply. But even in those cases were the cattle are moved, they are not moved far and often return to the area from which they were removed within a few days.
Passive season-long grazing is not compatible with maintaining water quality and wildlife habitat on Northern California national forests. That’s why the Project to Reform Public Land Grazing in Northern California advocates that Forest Service managers require modern grazing methods.
Because virtually all Northern California grazing allotments contain several distinct and separate pastures, it is entirely feasible to implement modern rest-rotation grazing methods. Where pastures are not sufficiently isolated, drift fences can be used to prevent cattle from returning to preferred areas. Implementing modern grazing technologies would spread impacts more evenly across the allotments and would result in improved water quality and less damage to riparian areas and wetlands.
Unfortunately, Forest Service managers have ignored the Project’s documentation of the serious environmental degradation that occurs as a result of “passive continuous grazing.” Responsible Forest Service officials fail to require the regular range riding and modern grazing methods which are essential if environmental degradation is going to be reduced. As the situation in early 2016 on the Malheur Refuge in Oregon and the earlier standoff with the Bundy Rancher Militia in Nevada demonstrate, when public land managers require good management (or even the payment of grazing fees) they are often subjected to threats, intimidation and worse. To avoid conflict, most public land grazing managers prefer to ignore the impacts of poorly managed grazing in order to avoid confrontation with ranchers and their radical supporters.
The situation has become so bad on the Klamath National Forest that Forest Service managers no longer require ranchers to remove their cattle from the range when the grazing permit says they should be removed. A growing number of those ranchers now do not even go out to gather their cattle and move them to the home ranch at the end of the grazing season. Instead they allow the cattle to walk home on their own, grazing on national forest land along the way for up to a month after those cattle were supposed to be removed from national forest land.
A steer on the Klamath National Forest walking home 10 days after all Cattle ere supposed to have been removed from public land
Forest Service managers are directed to charge excess use fees when a grazing permit holder allows his or her cattle to remain on national forest land after the date by which the grazing permit says all livestock are supposed to have been removed from national forest land. After observing that some permit holders on the Klamath National Forest don’t even begin gathering their cattle until the official off date and that others allowed their cattle to wander home on national forest land for up to a month after the permit said they should be removed, the Project used the Freedom of Information Act to obtain billing records for excess use charges. We wanted to see if grazing managers were collecting the extra grazing fees officially required.
Over the course of the past ten years, managers of the Klamath National Forest issued only one bill for excess use when a rancher allowed his cattle to remain on the range all winter. For most KNF grazing permit holders, however, there are never consequences, financial or otherwise, when they ignore grazing permit requirements; No wonder several don’t remove their cattle by the off date. Unfortunately, it is the land, wildlife and water quality which suffer as a result, while for the Forest Service managers who fail to do their duty there are also no consequences. The future
For six years the Project to Reform Public Land Grazing in Northern California has been meticulously documenting the lack of responsible management on Forest Service grazing allotments and the resulting environmental degradation. For six years we’ve written reports presenting what we found on-the-ground and calling on Forest Service managers to correct management deficiencies. Above all else, we have asked responsible Forest Service officials, the district rangers, to require regular herding that moves cattle to lessen the negative impacts of grazing. The response has been profoundly disappointing.
The failure of Forest Service managers to reform grossly inadequate and retrograde grazing management has caused the Project and its sponsors to adopt a new approach. We will now ask the superiors of those who mismanage grazing and those whose job is oversight of the Forest Service to correct the mismanagement. And we will begin to challenge the reauthorization of grazing on national forest land unless modern grazing methods are required.
The Project does not seek to end national forest grazing. What we seek is proper and responsible grazing management. If ranchers want to continue grazing their cattle on public land, they need to properly manage their livestock using modern methods. If ranchers are not willing to manage grazing properly, they have no business grazing public land.
In 2016 we will also seek to expand on-the-ground public land grazing monitoring to more national forests and to other public lands in Northern California. But that will depend on whether there are volunteers willing to go out on the grazing allotments to document conditions and how the cattle are being managed, if at all. As always, the Project is on the lookout for volunteers and interns who we will train to do the monitoring work Forest Service managers refuse to do.
Throughout this report I have referred to “poorly managed” national forest grazing. In reality, grazing on the national forest grazing allotments the Project has monitored is not just “poorly managed” it is virtually unmanaged. The project is determined to change that reality.
Cattle in need of management: a common sight on Northern California’s national forests www.nature.com/scientificreports
OPEN Human behaviour can trigger large carnivore attacks in developed countries received: 27 July 2015 Vincenzo Penteriani1,2, María del Mar Delgado2,3, Francesco Pinchera4, Javier Naves1, accepted: 06 January 2016 Alberto Fernández-Gil1, Ilpo Kojola5, Sauli Härkönen6, Harri Norberg6, Jens Frank7, Published: 03 February 2016 José María Fedriani1,8, Veronica Sahlén9,10, Ole-Gunnar Støen9, Jon E. Swenson9,10, Petter Wabakken11, Mario Pellegrini4, Stephen Herrero12 & José Vicente López-Bao2,7
The media and scientific literature are increasingly reporting an escalation of large carnivore attacks on humans in North America and Europe. Although rare compared to human fatalities by other wildlife, the media often overplay large carnivore attacks on humans, causing increased fear and negative attitudes towards coexisting with and conserving these species. Although large carnivore populations are generally increasing in developed countries, increased numbers are not solely responsible for the observed rise in the number of attacks by large carnivores. Here we show that an increasing number of people are involved in outdoor activities and, when doing so, some people engage in risk-enhancing behaviour that can increase the probability of a risky encounter and a potential attack. About half of the well-documented reported attacks have involved risk-enhancing human behaviours, the most common of which is leaving children unattended. Our study provides unique insight into the causes, and as a result the prevention, of large carnivore attacks on people. Prevention and information that can encourage appropriate human behaviour when sharing the landscape with large carnivores are of paramount importance to reduce both potentially fatal human-carnivore encounters and their consequences to large carnivores.
During the last few decades, large carnivore attacks on humans in developed countries have increased over time1–8 (Fig. 1). This is expected to increase people’s apprehension and reduce their willingness to share the landscape with large carnivores. Unfortunately, such rare events are usually overplayed by the media. Indeed, media cover- age of such attacks generally includes sensational texts and dreadful pictures (Extended Data 1), appealing more to the public’s emotions than their logic. Denominator neglect9 is a well-studied phenomenon leading humans to overestimate the risk of rare events that evoke strong emotions. Overestimating the risk of large carnivore attacks on humans irrationally enhances human fear and triggers a vicious cycle that may affect the increasingly posi- tive conservation status of many of these contentious species10–12. With an increasing number of large carnivore attacks on humans there is, now more than ever, a need for objective and accurate information regarding not only the long-term trend and underlying mechanisms of large carnivore attacks on humans, but also potentially risky situations and risk-enhancing human behaviours8. Surprisingly, the few available studies focus on attacks by
1Department of Conservation Biology, Estación Biológica de Doñana, C.S.I.C., c/Américo Vespucio s/n, 41092 Seville, Spain. 2Research Unit of Biodiversity (UMIB, UO-CSIC-PA), Oviedo University-Campus Mieres, 33600 Mieres, Spain. 3Metapopulation Research Centre, University of Helsinki, FI-00014 Helsinki, Finland. 4C.I.S.D.A.M., Via S. Liberata 1, Rosello (CH) I-66040, Italy. 5Natural Resources Institute Finland, P.O. Box 16, FI-96301 Rovaniemi, Finland. 6Finnish Wildlife Agency, Sompiontie 1, FI-00730 Helsinki, Finland. 7Grimsö Wildlife Research Station, Department of Ecology, Swedish University of Agricultural Sciences, 73091 Riddarhyttan, Sweden. 8Centre for Applied Ecology “Prof. Baeta Neves”, Institute Superior of Agronomy, University of Lisbon, Tapada da Ajuda, 1349-017 Lisboa, Portugal. 9Department of Ecology and Natural Resource Management, Norwegian University of Life Sciences, Postbox 5003, NO-1432 Ås, Norway. 10The Norwegian Environment Agency, P.O. Box 5672 Sluppen, N-7485 Trondheim, Norway. 11Faculty of Applied Ecology and Agricultural Sciences, Hedmark University College, Evenstad, NO-2480, Koppang, Norway. 12Faculty of Environmental Design, University of Calgary, Calgary, Alberta, Canada T2T 2Y2. Correspondence and requests for materials should be addressed to V.P. (email: [email protected])
Scientific Reports | 6:20552 | DOI: 10.1038/srep20552 1 www.nature.com/scientificreports/
Figure 1. Temporal trends in large carnivore attacks on humans in developed countries. The number of attacks on humans by large carnivores has increased significantly (Extended Data Table 1) during the last few decades for almost all large carnivores. The left panel shows the relationship between the number of large carnivore attacks in the US and the number of visitors (hundreds of millions, red line) in American protected areas since 1955, which has increased significantly over time (Extended Data Table 4). The right panels show (from top to bottom) the temporal trends in large carnivore attacks in Canada, as well as the trends of polar bear (Europe, Russia, the United States and Canada) and European brown bear (Sweden, Finland and Spain) attacks. It is worth noting that: (i) conflicts with polar bears have been increasing in the last decade. Causal factors include a growing human population and more tourists visiting polar bear areas, increased oil and gas development along the Arctic coastline, and decreasing ice volume and seasonal extent due to climate change63. Indeed, human-polar bear encounters are expected to increase as the sea ice continues to melt and hungry bears are driven ashore (http://www.polarbearsinternational.org/about-polar-bears/essentials/attacks-and- encounters; http://www.theguardian.com/world/2013/nov/04/polar-bear-attacks-scientists-warn-warming- arctic); (ii) the remarkable increase in coyote attacks may be related to both the recent substantial expansion of the coyote range in eastern North America64 and increased conflicts in suburban residential areas. In these areas, coyotes can relax human avoidance mechanisms as a result of relying on anthropogenic food resources and even intentional feeding by residents4; and (iii) wolves were the only species to show a decreasing trend in the number of attacks, declining from 10 attacks during the decade 1975–1984 to only two or three attacks per decade starting in 1985. (The brown bear picture has been downloaded from 123RF ROYALTY FREE STOCK PHOTOS (http://www.123rf.com), Image ID 7250879, Eric Isselee).
single carnivore species and thus they do not provide a comprehensive perspective concerning the pervasiveness and socio-ecological correlates of this phenomenon in developed countries. Our main hypothesis is that lack of knowledge of people about how to avoid risky encounters with large car- nivores engenders risk-enhancing behaviours, which can determine an increase in the number of attacks if more humans are sharing landscape with large carnivores. Three main predictions arise from this hypothesis: (1) an increased number of people are engaging in outdoor leisure activities in areas inhabited by large carnivores; (2) many people are not prepared to safely enjoy outdoor activities or they behave inappropriately in the countryside; and (3) large carnivore attacks are influenced by the interaction between several human- and animal-related factors. Thus, we first explored whether the long-term patterns in the number of attacks have been similar among dif- ferent large carnivore species, and how they varied throughout the year. Then, we evaluated whether there might have been a general long-term change in the attack patterns by assessing whether victim ages and the frequency of attacks on parties vs. lone humans have changed in a congruent manner for the different species. Finally, we assessed the possible relationships between temporal trends of attacks on humans and outdoor activities, as well as the role that risk-enhancing human behaviour can have played in the observed increase in large carnivore attacks.
Scientific Reports | 6:20552 | DOI: 10.1038/srep20552 2 www.nature.com/scientificreports/
Figure 2. Temporal trends in large carnivore attacks on humans in developed countries: monthly patterns. Most large carnivore attacks occurred from late spring to early autumn, when most people usually engage in outdoor activities. (The coyote picture has been downloaded from 123RF ROYALTY FREE STOCK PHOTOS (http://www.123rf.com), Image ID 14988151, James Mattil).
We analysed the circumstances of ca. 700 large carnivore attacks on people from 1955 in detail (when more reliable data became available) until the present for six species responsible for most of the large carnivore attacks recorded in North America and Europe: brown/grizzly bear Ursus arctos, black bear Ursus americanus, cougar Puma concolor, wolf Canis lupus and coyote Canis latrans in North America, brown bear in Europe, and polar bear Ursus maritimus circumpolar, i.e. Europe, Russia and North America. We also collected statistics concerning outdoor activities during the same period (see Methods for more details). The number of large carnivore attacks on people has increased significantly over time, with contrasting trends across species (Fig. 1; Extended Data Table 1). In North America, coyotes (31.0% of the total number of attacks) and cougars (25.7%) were responsible for the majority of attacks, followed by brown bears (13.2%), black bears (12.2%) and wolves (6.7%). A similar increase over time was observed for brown bears (9.3%) in Europe and cir- cumpolar for polar bears (1.9%; Fig. 1). Moreover: (a) the age of victims has also increased significantly over time showing different patterns across species (Extended Data Table 2 and Extended Data Fig. 1); and (b) the propen- sity to attack lone humans or parties depends on the large carnivore species, and only a slight but non-significant increasing trend of attacks on parties has been observed (Extended Data Table 3 and Extended Data Fig. 2A,B). The patterns of attacks reported here may also reflect an increasing number of bold individuals in large carni- vore populations, as this trait is often correlated with aggressiveness13,14, and this might lead to more aggressive responses when large carnivores encounter humans. We hypothesise that intense and prolonged human-caused mortality imposes selection pressures on target populations (selective removal of certain phenotypes) and might lead to rapid evolutionary changes15. Natural selection maintains a mix of behavioural phenotypes in popula- tions16, the shy-bold behavioural continuum17; bold individuals thrive on risk and novelty, whereas shy individu- als shrink from the same situations18. Persecution, however, is expected to result in the disproportionate removal of bold individuals, as they are less cautious19, and thus more likely to be killed. As a consequence, shy individ- uals might have been overrepresented in remnant large carnivore populations in the past17,18,20–22. Additionally, individuals may become more vigilant and actively avoid contact with humans during times of intense persecu- tion23. Although the history of large carnivore persecution and conservation differ across regions9, the contempo- rary conservation paradigm emerged during the 1960s–1970s24, when most bounty systems were banned25 and large carnivores were reclassified from vermins or bountied predators to game or protected species. Since then, although large carnivores have continued to be hunted or managed (Extended Data Fig. 3), most populations have generally increased during the past four decades9,11,12. Increasing population trends in conjunction with relaxed artificial selection may potentially engender higher variation in behavioural temperaments26, which is likely to alter individual responses to human encounters22. This significant increase of large carnivore populations in both North America and Europe, and their consequent range expansion, also may contribute to explain the observed increase in the attacks on humans. However, similar to the increasing trend in attacks, the number of people engaging in outdoor leisure activities also has risen over time, a phenomenon that is significantly correlated with the observed trend in the number of attacks (Fig. 1; Extended Data Table 4, Extended Data Fig. 4A–C). Seasonally, most of the attacks occurred between late spring and early autumn (Fig. 2), when most people pursue outdoor activities7,8; in addition, because bears hibernate, they are unlikely to attack people in winter. Such an increase in recreational activities in areas inhabited by large carnivores implicitly increases the probability of a risky encounter and, therefore, a potential attack. However, even with more people visiting those areas, attacks are still extremely rare (Fig. 1): although some people may only focus on the total number of attacks, we have to bear in mind the long time period during which these attacks occurred.
Scientific Reports | 6:20552 | DOI: 10.1038/srep20552 3 www.nature.com/scientificreports/
Figure 3. The number fo attacks is modulated by human behaviour. Around half of the attacks were associated with risk-enhancing human behaviours. Out of 271 well-documented attacks, 47.6% were associated with certain human behaviours that may have contributed to the probability of suffering an attack. Within the principal category (children left unattended by their parents), the main species responsible for 91.9% of these attacks were cougars (50.8%), coyotes (27.9%) and black bears (13.2%). (The cougar picture has been downloaded from 123RF ROYALTY FREE STOCK PHOTOS (http://www.123rf.com), Image ID 2597979, Eric Isselee).
Remarkably, risk-enhancing human behaviour has been involved in at least half of the well-documented attacks (47.6%; Fig. 3). From highest to lowest, the five most common human behaviours occurring at the time of an attack were (a) parents leaving children unattended, (b) walking an unleashed dog, (c) searching for a wounded large carnivore during hunting, (d) engaging in outdoor activities at twilight/night and (e) approaching a female with young. These are clearly risk-enhancing behaviours when sharing the landscape with large carnivores. For example, the most frequently recorded human behaviour was children left unattended (47.3%), which were most often attacked by cougars (50.8% of the attacks), coyotes (27.9%) and black bears (13.2%). Risk-enhancing human behaviour is not the sole reason behind large carnivore attacks on humans. The causes of the other half of the attacks do not seem to be related to risk-enhancing human behaviour, for example, accidentally walking close to a mother with young or to a carcass with a bear nearby or an encounter with a food-conditioned individual (which is an indirect result of a risk-enhancing human behaviour8). Thousands of interactions occur between people and large carnivores with no human injuries or fatalities. Even if attacks have increased over time, they remain extremely rare events (e.g. a cross-continental average of 24.1 attacks and 3.9 fatalities per year during the last decade, all species pooled; Fig. 1). Other wildlife (bees and mosquitos, spiders, snails, snakes and ungulates) and domestic dogs are far more responsible for human fatali- ties1,27. But humans are not the only victims. When attacks occur, large carnivores are frequently killed and nega- tive attitudes towards large carnivores harden6. Lethal removal of ‘problematic’ individuals is effective in solving the local problem caused by a given individual28, but generally this happens after an aggressive behaviour, human injury or death has occurred. Consequently, both humans and carnivores suffer from these incidents. After decades of minimal interaction between humans and large carnivores in many regions of developed countries, many people involved in outdoor activities may lack knowledge about how to avoid risky encounters with large carnivores and what to do when such encounters occur. From an early age most of us learn social norms, rules and how to decrease risks in urban environmental settings, but much less effort is expended to teach us how to safely enjoy outdoor activities or to behave appropriately in the countryside. However, it is up to us to reduce the likelihood of an attack. The increasing human presence in areas inhabited by large carni- vores, together with their population recoveries9,11,12, requires an improvement in information, education and prevention guidelines, and their enforcement, which are of paramount importance to reduce both the risks to humans and the killing of carnivores1,4,7,28,29. Educating people that share landscape with large carnivores can represent a crucial factor to help reducing the number of attacks and also the negative attitudes towards large carnivore conservation, especially because of the difficulty to envisage risk estimates. Indeed, scenarios of attacks are extremely different and may depend on many different factors, such as human population and carnivore densities, time of the day, human activities, personality and condition of the large carnivore, party size or even subtle details, like the presence of an unleashed dog at the moment of the attack and/or the landscape features of the area where an attack has happened. As conflicts between humans and large carnivores continue to increase, accurate information becomes crucial to informed human–wildlife conflict management. Communicating about large carnivore-inflicted human injuries and fatalities in a statistical manner contributes to better understanding of common patterns in large carnivore attacks, further reduces chances of injury or death and promotes public appreciation of these species. An important strategy to reduce attacks on humans is to inform people how to avoid
Scientific Reports | 6:20552 | DOI: 10.1038/srep20552 4 www.nature.com/scientificreports/
and manage aggressive encounters. But nowadays, educational and interpretive efforts aimed at decreasing the risk of large carnivore attacks should not focus exclusively on people living in rural and wilderness areas. Indeed, many people living in cities should also be included within the category of groups at risk because of the increasing number of them enjoying outdoor activities in areas inhabited by large carnivores and the expanding population of carnivores (mainly coyotes) in suburban areas. Although large carnivore attacks on humans are influenced by the interaction between multiple human- and animal-related factors, adapting our own behaviours when coexisting with large carnivores has the potential to reduce the number of attacks to about half of today’s level. The examples provided by the numerous cases of chil- dren injured/killed while left unattended by their parents, attacks on people jogging/walking alone at twilight and during hunting, should make us reflect on our responsibilities, the possibility of decreasing the number of these tragic events and changing the observed trends. Understanding the circumstances associated with large carnivore attacks should help us to reduce them and thereby minimize the role that fear and supposition may play in large carnivore management and conservation. Methods Collection of records of large carnivore attacks on humans. Records of large carnivore attacks (i.e. attacks resulting in physical injury or death) on humans for the brown bear, black bear, cougar, wolf and coyote were collected for North America (the United States and Canada). In addition, with the aim to broaden our research and obtain a general picture of large carnivore attacks on humans in developed countries, we com- plemented the North American dataset with information on brown bear attacks in three European countries (Sweden, Finland and Spain) as well as data on attacks by polar bears in Europe (Svalbard; Norway), Russia, the United States, and Canada. Our time period spanned from 1955 to 2014 and our search resulted in a total of 697 attacks of large carnivores on people. We consider that we both recorded the majority of such events occurring during the last six decades in these developed countries and avoided bias due to possible changes in reporting probability given (i) the large number of experienced people involved in the work (some of them had their own database on attacks, which started at the beginning of the 1900s), (ii) the multiple sources of information used to collect recorded attacks and (iii) the sen- sational nature and media impacts of attacks that end with injuries or the death of the victim since the beginning of the past century. Records of attacks were collected from unpublished reports and PhD/MS theses, webpages (last accessed in November 2014, but currently available at the specific addresses listed by species below), books and scientific articles, as well as personal datasets from some of the co-authors. In addition, to complete the data obtained from the above-cited sources, we also collected dozens of news reports from online newspapers. To do this, for each species and area, we searched on an annual basis for news articles on Google using the combination of the following terms: “species name” + “attack” and “species name” + “attack” + “human”. Because of the use of multiple sources, several attacks recurred repeatedly during the search, but we used information such as date, locality and sex/age of the victims to prevent duplicate records in the dataset. However, the lack of some records would still not result in a bias in the general patterns we observed in the present work, because: (i) we followed the same procedure for each species and, thus, we collected at least an equally biased sample of attacks per species and (ii) patterns of attacks on humans over time are less sensitive to unequally biased samples of attacks than quantitative comparisons of the frequency of attacks across species (which is not the aim of the present work). When possible, we recorded the following information for each attack: (1) species; (2) year; (3) month; (4) country; (5) time of the attack during the day (which we classified into three categories: twilight, day, night); (6) activity of the victim (15 categories: hunting, fishing, field work, camping, hiking, jogging, skiing, biking, horse riding, fruit/mushroom picking, photography, walking, dog walking, activity near the house/in the backyard, playing); (7) size of party being attacked (simplified into three categories: victim alone, child – from 0 to 16 years old– in a party of adults, adult – > 16 years old – in a party of adults); (8) end of the attack, i.e. attack resulting in human injuries or death; and (9) scenario when the attack occurred, i.e. the factor that could have triggered the attack. We were able to delineate eight categories: female with young, aggressive reaction after a sudden encoun- ter (i.e. a person surprises the large carnivore at close range), food defence (e.g. a bear close to a carcass), food conditioning (i.e. encounter with a large carnivore that consumes human-derived foods, consequently associating people with easily accessible, attractive foods, and which has lost much of its avoidance mechanisms towards humans), predatory (i.e. when the large carnivore exploited a human as prey), wounded animal (i.e. during hunt- ing), feeding large carnivores, and presence of one or more dogs. Unleashed dogs can exacerbate the probability of a large carnivore attack, because a dog that runs away from a large carnivore towards the owner can trigger a dangerous situation when the carnivore chases it30. When dogs were involved, large carnivores usually focused their attention on the dog rather than on the person. However, in some instances the human was attacked as a consequence of its proximity to the dog or because of its reaction towards the large carnivore. Below, we describe the sources used to collect data on large carnivore attacks on people for each species since 1955:
1. North American brown and black bears. We recorded a total of 92 and 85 attacks, respectively. Informa- tion was compiled from28, Wikipedia List of fatal bear attacks in North America (http://en.wikipedia.org/ wiki/List_of_fatal_bear_attacks_in_North_America), Fatal Bear Attack Statistics for the USA & Canada (http://www.blackbearheaven.com/bear-attack-statistics.htm) and online newspapers. Additionally, we also extracted information for the black bear from8 and California Black Bear Public Safety Incidents, California Department of Fish and Wildlife (https://www.wildlife.ca.gov/News/Bear/Bear-Incidents). 2. Cougar. We recorded a total of 179 attacks. Data on attacks were collected from1,31, the List of Mountain Lion Attacks (http://www.cougarinfo.org/attacks.htm), Mountain Lion Attacks from 1991 to 2000 (http:// www.cougarinfo.org/attacks2.htm), Mountain Lion Attacks from 2001 to 2010 (http://www.cougarinfo.org/
Scientific Reports | 6:20552 | DOI: 10.1038/srep20552 5 www.nature.com/scientificreports/
attacks3.htm), Mountain Lion Attacks from 2011 to Now (http://www.cougarinfo.org/attacks4.htm) and on- line newspapers. 3. Wolf. We recorded a total of 47 attacks. Data on attacks were collected from32,33, the Wikipedia List of wolf attacks in North America (http://en.wikipedia.org/wiki/List_of_wolf_attacks_in_North_America), Wikipe- dia List of wolf attacks (http://en.wikipedia.org/wiki/List_of_wolf_attacks), Wolf Attacks on Humans (http:// www.aws.vcn.com/wolf_attacks_on_humans.html) and online newspapers. We did not include the wolf in Europe, because predatory attacks on people have been extremely rare during the last six decades, with the last recorded predatory attack occurring in 1974 in Spain32. 4. Coyote. We recorded a total of 216 attacks. Data on attacks were collected from4,34–36, the Wikipedia Coyote attacks on humans (http://en.wikipedia.org/wiki/Coyote_attacks_on_humans), Coyote Attacks on Children (http://www.varmintal.com/attac.htm), Coyote Attacks: An Increasing Suburban Problem (http://escholar- ship.org/uc/item/8qg662fb), Coyote Attacks On People in the U.S. and Canada (http://tchester.org/sgm/lists/ coyote_attacks.html) and online newspapers. 5. European brown bear. We recorded a total of 65 attacks. Information from Spain was available from the un- published personal database of J.N. and A.F.G., whereas Fennoscandian records were obtained from37,38 and unpublished data from I.K., H.N. and J.F. 6. Polar bear. We recorded a total of 13 attacks. Information was recorded from Wikipedia List of fatal bear attacks in North America (http://en.wikipedia.org/wiki/List_of_fatal_bear_attacks_in_North_America) and online newspapers (both North American and European –some attacks have been recently recorded in the Norwegian Svalbard archipelago–).
Collection of records on outdoor human activities. Data on outdoor activities was only available for the US and Sweden. We collected the following information: (1) annual recreation visitation in American Protected Areas published by the National Park Service Visitor Use Statistics (IRMA data system), National Park Service, U.S. Department of the Interior, Natural Resource Stewardship and Science (https://irma.nps. gov/Stats/Reports/National). To reduce bias in our analyses, we only used information from the National Parks located in the 30 states where at least one large carnivore attack occurred since 1955 (Alaska, Arizona, California, Colorado, Florida, Georgia, Idaho, Illinois, Kansas, Kentucky, Massachusetts, Michigan, Minnesota, Montana, New Hampshire, New Jersey, New Mexico, New York, North Carolina, North Dakota, Ohio, Oklahoma, Pennsylvania, Tennessee, Texas, Utah, Vermont, Virginia, Washington, Wyoming); (2) statistics on number of people doing outdoor activities in the US, which were obtained from39, the U.S. Department of the Interior, U.S. Fish and Wildlife Service, and U.S. Department of Commerce, U.S. Census Bureau40–44; (3) American trends in the sporting goods market related to outdoor activities associated with attacks (cross-country training shoes, jogging and running shoes, camping, optics, snow skiing, bicycles and related supplies), which were collected from the U.S. Census Bureau, Statistical Abstract of the United States45,46; and (4) Statistics Sweden’s time series tables concerning outdoor activities, which derive from ULF surveys (Living Conditions Surveys) from 1975 onwards (http://www.scb.se/sv_/Hitta-statistik/Statistik-efter-amne/Levnadsforhallanden/Levnadsforhallanden/ Undersokningarna-av-levnadsforhallanden-ULFSILC/12202/12209/#). Information on outdoor human activities was only used to support the highlighted trends and patterns of large carnivore attacks; thus, its sole function is to be supportive to the main text and it was not used in our analyses.
Collection of records on large carnivore harvest. We used data from brown bear, black bear, cougar and wolf harvests in certain US and Canadian states as examples of trends and numbers in large carnivore harvest over time. First, brown bear harvesting records for Alaska and British Columbia were obtained from the Alaska Department of Fish and Game47,48 and M. Wolowicz unpublished data (Big Game Harvest Statistics 1976–2012, British Columbia), respectively. Second, data on black bear harvesting statistics in Alaska was obtained from the Alaska Department of Fish and Game49–51. Third, cougar harvesting records in Colorado, Alberta and British Columbia were obtained from52–54, J. Apker unpublished data (Colorado Division of Wildlife) and M. Wolowicz unpublished data (Big Game Harvest Statistics 1976–2012, British Columbia). Finally, wolf harvesting statistics were extracted from the Alaska Department of Fish and Game55–58. Again, as we did for the information on out- door human activities, the records on large carnivore harvest were only used to support the highlighted trends and patterns of large carnivore attacks; thus, their sole function is to be supportive to the main text and they were not used in our analyses.
Data analysis. Considering the total dataset on large carnivore attacks since 1955, we first assessed whether the number of attacks varied over time, on a yearly basis, and among species by fitting a Generalized Linear Model (GLM) with the number of attacks against year and species (Extended Data Table 1). We also included the inter- action term between year and species to account for the fact that the number of attacks may vary over time het- erogeneously across species. Because our data were overdispersed, we fitted the GLM using a Negative Binomial distribution instead of a Poisson distribution. Next, to assess a potential change in the behavioural temperament of large carnivores over time, we tested whether the log-transformed age of the victim and party size (three lev- els) varied over time and among species by fitting a linear model with a Gaussian distribution and a GLM with a multinomial distribution (three levels), respectively (Extended Data Tables 2 and 3). Party size was classified into three categories, which allows differentiating between attacks on lone individuals and groups, as well as if the vic- tim in a group was a young person: i) the victim was alone; ii) the victim was a young person (< 16 years old) in a group of adults (2 or more people); and iii) the victim was an adult (> 16 years old) in a group of adults (2 or more people). We also considered the interaction term in these models to account for the fact that the surrogates of the changes in the temperament of large carnivores used may vary over time differently across species. Finally, we
Scientific Reports | 6:20552 | DOI: 10.1038/srep20552 6 www.nature.com/scientificreports/
analysed a subset of the dataset considering only those attacks occurring in the US and, together with information on human influx in natural areas, we tested if the number of attacks was related to the number of people involved in outdoor activities by building a GLM with a Gamma distribution, considering year, the number of visitors and their interaction term as factors in the model (Extended Data Table 4). For each analysis, we used an information theoretic framework to rank a set of competing models based on AIC (Akaike’s Information Criterion [AIC]59). We used a stepwise selection procedure to create a candidate set of a priori competing models starting from the simplest null model (intercept only model) to the full model (Extended Data Tables 1, 2, 3 and 4). To select the best candidate model, we used AIC value corrected for small sample sizes (AICc) and Weighted AIC, which indicates the probability that the model selected is the best among the candidates59. Models within Δ AIC < 2 were considered to have substantial empirical support59. All statisti- cal analyses were performed using R 3.0.2 statistical software60. GLMs were run with the “lme4”61 and “nlme”62 package. References 1. Beier, P. Cougar attacks on humans in the United States and Canada. Wildl. Soc. Bull. 19, 403–412 (1991). 2. National Geographic.com. Predator Attacks Escalate as Americans Encroach on Wildlife Habitat. http://news.nationalgeographic. com/news/2001/08/0827_wirepredators.html (2001) (Date of access: 24/09/2015). 3. Herrero, S. & Higgins, A. Human injuries inflicted by bears in Alberta: 1960–98. Ursus 14, 44–54 (2003). 4. Timm, R. M., Baker, R. O., Bennett, J. R. & Coolahan, C. C. Coyote attacks: an increasing suburban problem. Trans. No. Amer. Wildlife & Nat. Resources Conf. 69, 67–88 (2004). 5. CNN.com. Predators, humans sharing habitats a volatile mix. http://www.cnn.com/2006/TECH/science/05/30/animal.attacks/ index.html (2006). (Date of access: 24/09/2015). 6. Conover, M. R. Why are so many people attacked by predators? Hum. Wildl. Conflicts 2, 139–140 (2008). 7. White, L. A. & Gehrt, S. D. Coyote attacks on humans in the United States and Canada. Hum. Dimensions Wildl. 14, 419–432 (2011). 8. Herrero, S., Higgins, A., Cardoza, J. E., Hajduk, L. I. & Smith, T. S. Fatal Attacks by American Black Bear on People: 1900–2009. J. Wildl. Manage. 75, 596–603 (2011). 9. Reyna, V. F. How people make decisions that involve risk. A dual-processes approach. Curr. Dir. Psychol. Sci. 13, 60–66 (2004). 10. Chapron, G. et al. Recovery of large carnivores in Europe’s modern human-dominated landscapes. Science 346, 1517–1519 (2014). 11. Eberhardt, L. L. & Breiwick, J. M. Trend of the Yellowstone grizzly bear population. Int. J. Ecol., doi: 10.1155/2010/924197 (2010). 12. LaRue, M. A. et al. Cougars are recolonizing the Midwest: analysis of cougar confirmations during 1990–2008. J. Wildl. Manage. 76, 1364–1369 (2012). 13. Moya-Laraño, J. Genetic variation, predator–prey interactions and food web structure. Phil. Trans. R. Soc. B 366, 1425–1437 (2011). 14. Wolf, M. & Weissing, F. J. Animal personalities: consequences for ecology and evolution. Trends Ecol. Evol. 27, 452–461 (2012). 15. Allendorf, F. W. & Hard, J. J. Human-induced evolution caused by unnatural selection through harvest of wild animals. Proc. Natl. Acad. Sci. USA 106, 9987–9994 (2009). 16. Schreiber, S. J., Bürger, R. & Bolnick, D. I. The community effects of phenotypic and genetic variation within a predator population. Ecology 92, 1582–1593 (2011). 17. Smith, B. R. & Blumstein, D. T. Fitness consequences of personality: a meta-analysis. Behav. Ecol. 19, 448–455 (2008). 18. Wilson, D. S., Clark, A. B., Coleman, K. & Dearstyne, T. Shyness and boldness in humans and other animals. Trends Ecol. Evol. 9, 442–446 (1994). 19. Bremner-Harrison, S., Prodohl, P. A. & Elwood, R. W. Behavioural trait assessment as a release criterion: boldness predicts early death in a reintroduction programme of captive-bred swift fox (Vulpes velox). Anim. Conserv. 7, 313–320 (2004). 20. Biro, P. A. & Post, J. R. Rapid depletion of genotypes with fast growth and bold personality traits from harvested fish populations. Proc. Natl. Acad. Sci. USA 105, 2919–2922 (2008). 21. Ciuti, S. et al. Human selection of elk behavioural traits in a landscape of fear. Proc. R. Soc. B 279, 4407–4416 (2012). 22. Madden, J. R. & Whiteside, M. A. Selection on behavioural traits during ‘unselective’ harvesting means that shy pheasants better survive a hunting season. Anim. Behav. 87, 129–135 (2014). 23. Shivik, J. The Predator Paradox: Ending the War with Wolves, Bears, Cougars, and Coyotes (Beacon Press 2014). 24. Kellert, S. R., Black, M., Rush, C. R. & Bath, A. J. Human Culture and Large Carnivore Conservation in North America. Conserv. Biol. 10, 977–990 (1996). 25. Schwartz, C. C., Swenson, J. E. & Miller, S. D. Large carnivores, moose, and humans: a changing paradigm of predator management in the 21st century. Alces 39, 41–63 (2003). 26. Dall, S. R. X., Houston, A. I. & McNamara, J. M. The behavioural ecology of personality: consistent individual differences from an adaptive perspective. Ecol. Lett. 7, 734–739 (2004). 27. http://www.statista.com/chart/2203/the-worlds-deadliest-animals/ (Date of access: 24/09/2015). 28. Herrero, S. Bear attacks. Their causes and avoidance (Lyons Press, 2002). 29. Løe, J. & Røskaft, E. Large carnivores and human safety: a review. Ambio 33, 283–288 (2004). 30. Hristienko, H. & Herrero, S. Are dog “saviours” or are they contributing factors in black bear attacks on people? Int. Bear News 23, 19 (2014). 31. Colorado Division of Wildlife. Reported Lion Attacks on Humans, 1990 to present, report n° 80216-1000, Denver, Colorado (2011). 32. Linnell, J. et al. The fear of wolves: A review of wolf attacks on humans. NINA Oppdragsmelding 731 (2002). 33. McNay, M. E. A case history of wolf–human. Encounters in Alaska and Canada. Alaska Department of Fish and Game, Wildlife Technical Bulletin 13 (2002). 34. Carbyn, L. N. Coyote attacks on children in western North America. Wildl. Soc. Bull. 172, 444–446 (1989). 35. Hsu, S. S. & Hallagan, L. F. Case report of a coyote attack in Yellowstone National Park. Wild. Environ. Med. 2, 170–172 (1996). 36. Wildlife Damage Management Working Group of The Wildlife Society In: Proceedings 12th Wildlife Damage Management Conference (eds Nolte, D. L., Arjo, W. M. & Stalman, D. H.) 272–286 (2007). 37. De Giorgio, F., Rainio, J., Pascali, V. L. & Lalu, K. Bear attack – A unique fatality in Finland. Forensic Sci. Int. 173, 64–67 (2007). 38. Sahlén, V. Encounters between brown bears and humans in Scandinavia – contributing factors, bear behavior and management perspectives. PhD Thesis, Norwegian University of Life Sciences, Norway (2013). 39. Cordell, H. K. et al. Outdoor recreation in American life: a national assessment of demand and supply trends. Sagamore Publishing 219–321 (1999). 40. US Department of the Interior, Fish and Wildlife Service and US Department of Commerce, Bureau of the Census. National Survey of Fishing, Hunting, and Wildlife-Associated Recreation (1991). 41. US Department of the Interior, Fish and Wildlife Service and US Department of Commerce, Bureau of the Census. National Survey of Fishing, Hunting, and Wildlife-Associated Recreation (1996). 42. US Department of the Interior, Fish and Wildlife Service and US Department of Commerce, US Census Bureau. National Survey of Fishing, Hunting, and Wildlife-Associated Recreation (2001).
Scientific Reports | 6:20552 | DOI: 10.1038/srep20552 7 www.nature.com/scientificreports/
43. US Department of the Interior, Fish and Wildlife Service, and US Department of Commerce, US Census Bureau. National Survey of Fishing, Hunting, and Wildlife-Associated Recreation (2006). 44. US Department of the Interior, US Fish and Wildlife Service, and US Department of Commerce, US Census Bureau. National Survey of Fishing, Hunting, and Wildlife-Associated Recreation (2011). 45. US Census Bureau, Statistical Abstract of the United States. The Sporting Goods Market in 2002 and prior issues. Section 26: Arts, Entertainment, and Recreation, National Sporting Goods Association, Mt. Prospect, pp. 767–786 USA (2003). 46. US Census Bureau, Statistical Abstract of the United States. The Sporting Goods Market in 2010 and prior issues. Section 26: Arts, Entertainment, and Recreation. National Sporting Goods Association, Mt. Prospect, pp. 757–778, USA (2012). 47. Alaska Department of Fish and Game. Brown bear management report of survey-inventory activities 1 July 2002–30 June 2004 (Juneau, Alaska 2005). 48. Alaska Department of Fish and Game. Brown bear management report of survey-inventory activities 1 July 2008–30 June 2010 (Juneau, Alaska 2011). 49. Alaska Department of Fish and Game. Black bear management report of survey-inventory activities 1 July 1998–30 June 2001 (Juneau, Alaska 2002). 50. Alaska Department of Fish and Game. Black bear management report of survey-inventory activities 1 July 2001–30 June 2004 (Juneau, Alaska 2005). 51. Alaska Department of Fish and Game. Black bear management report of survey-inventory activities 1 July 2007–30 June 2010 (Juneau, Alaska 2011). 52. Hornocker, M. & Negri, S. (eds). Cougar. Ecology and Conservation (University of Chicago Press, 2010). 53. Wainwright, C. J., Darimont, C. T. & Paquet, P. C. British Columbia’s Neglected Carnivore: a Conservation Assessment and Conservation Planning Guide for Cougars. Raincoast Conservation Foundation, Sidney BC (2010). 54. Knopff, K. H., Webb, N. F. & Boyce, M. S. Cougar population status and range expansion in Alberta during 1991–2010. Wildl. Soc. Bull, doi: 10.1002/wsb.369 (2013). 55. Alaska Department of Fish and Game. Wolf management report of survey-inventory activities. Federal Aid in Wildlife Restoration 1 July 1996–30 June 1999 (Juneau, Alaska 2000). 56. Alaska Department of Fish and Game. Wolf management report of survey-inventory activities 1 July 1999–30 June 2002 (Juneau, Alaska 2003). 57. Alaska Department of Fish and Game. Wolf management report of survey-inventory activities 1 July 2002–30 June 2005 (Juneau, Alaska 2006). 58. Alaska Department of Fish and Game. Wolf management report of survey-inventory activities 1 July 2008–30 June 2011 (Species Management Report, ADF&G/DWC/SMR-2012-4, Juneau, Alaska 2011). 59. Burnham, K. P. & Anderson, D. R. Model selection and multimodel inference: a practical information-theoretic approach (Springer- Verlag, 2002). 60. R Development Core Team. R: A language and environment for statistical computing (R Foundation for Statistical Computing, 2013). 61. Bates, D. M. & Sarkar, D. lme4: Linear-mixed-effects models using S4 classes. R package version 0, 99875–6 (2007). 62. Pinheiro, J., Bates, D., DebRoy, S. & Sarkar, D. and the R Core team. nlme: Linear and Nonlinear Mixed Effects Models. R Package Version 3, 1–96 (2009). 63. Herreman, J. Management of polar bears in northern Alaska communities. 4th International human-bear conflicts workshop summary, Missoula, Montana, 9–10 (2012). 64. Levy, S. The new top dog. Nature 485, 296–297 (2012). Acknowledgements We thank J. Apker and M. Wolowicz for harvest statistics for cougars in Colorado and brown bears and cougars in British Columbia, respectively, as well as K. Logan and A. Rodrigues for general data on carnivore hunting and attacks. P. Fredman, G. Nordström and D. Rydberg helped collect statistics on outdoor human activities in Sweden. Marco Festa-Bianchet, Brock Fenton and an anonymous referee improved the first draft of the manuscript. This work was supported by the Spanish Ministry of Economy and Competitiveness (CGL2012– 33240; FEDER co-financing). MMD received a grant from the Kone Foundation (#44-6977). JVLB was supported by the Spanish Ministry of Economy and Competitiveness. Author Contributions V.P., M.M.D., F.P. and J.V.L.B. initiated and conceived the study, V.P. prepared the database with the collaboration of M.M.D., J.V.L.B., J.N., A.F.-G., I.K., S.H. and H.N. M.M.D. and J.V.L.B. analysed the data, V.P. wrote the manuscript with M.M.D. and J.V.L.B. and all authors commented on the manuscript draft. Additional Information Supplementary information accompanies this paper at http://www.nature.com/srep Competing financial interests: The authors declare no competing financial interests. How to cite this article: Penteriani, V. et al. Human behaviour can trigger large carnivore attacks in developed countries. Sci. Rep. 6, 20552; doi: 10.1038/srep20552 (2016). This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
Scientific Reports | 6:20552 | DOI: 10.1038/srep20552 8 REVIEW
ECOLOGY 2015 © The Authors, some rights reserved; exclusive licensee American Association for ’ the Advancement of Science. Distributed Collapse of the world s largest herbivores under a Creative Commons Attribution 1 1,2 1 License 4.0 (CC BY). William J. Ripple, * Thomas M. Newsome, Christopher Wolf, 10.1126/sciadv.1400103 Rodolfo Dirzo,3 Kristoffer T. Everatt,4 Mauro Galetti,5 Matt W. Hayward,4,6 Graham I. H. Kerley,4 Taal Levi,7 Peter A. Lindsey,8,9 David W. Macdonald,10 Yadvinder Malhi,11 Luke E. Painter,7 Christopher J. Sandom,10 John Terborgh,12 Blaire Van Valkenburgh13
Large wild herbivores are crucial to ecosystems and human societies. We highlight the 74 largest terrestrial herbi- vore species on Earth (body mass >– 100 kg), the threats they face, their important and often overlooked ecosystem effects, and the conservation efforts needed to save them and their predators from extinction. Large herbivores are generally facing dramatic population declines and range contractions, such that ~60% are threatened with extinc- tion. Nearly all threatened species are in developing countries, where major threats include hunting, land-use change, and resource depression by livestock. Loss of large herbivores can have cascading effects on other species including large carnivores, scavengers, mesoherbivores, small mammals, and ecological processes involving vege- tation, hydrology, nutrient cycling, and fire regimes. The rate of large herbivore decline suggests that ever-larger swaths of the world will soon lack many of the vital ecological services these animals provide, resulting in enormous ecological and social costs.
INTRODUCTION and conservation action to help diminish the imminent possibility Terrestrial mammalian herbivores, a group of ~4000 species, live in of losing the remaining large herbivores from many ecosystems throughout every major ecosystem on Earth except Antarctica. Here, we consider the world. the 74 wild herbivore species with mean adult body masses ≥100 kg. These largest species represent four orders (Proboscidea, Primates, Ce- tartiodactyla, and Perissodactyla) and 11 families (Elephantidae, Rhino- STATUS cerotidae, Hippopotamidae, Giraffidae, Bovidae, Camelidae, Tapiridae, Equidae, Cervidae, Suidae, and Hominidae). Most of these species are According to the International Union for the Conservation of Nature entirely herbivorous, but some are generalists (for example, Suidae). (IUCN), 44 of the 74 largest terrestrial herbivores (~60%) are listed as Herein, we provide the first comprehensive review that includes the threatened with extinction (including 12 critically endangered or extinct endangerment status and key threats to the world’s largest herbivores in the wild), and 43 (~58%) have decreasing populations [(1); table S1]. (≥100 kg), the ecological consequences of their decline, and actions Their current population sizes exhibit large differences among species, needed for their conservation. We review how the combined impacts spanning over four orders of magnitude, with some populations estimated of hunting, encroachment by humans and their livestock, and habitat to comprise fewer than 100 individuals [for example, Javan rhinoceros loss could lead to the extinction of a suite of large herbivores relatively (Rhinoceros sondaicus)], whereas a few others [for example, Eurasian elk/ soon. By reviewing their ecological roles, we show how the loss of large moose (Alces alces)] comprise more than 1 million individuals (table S1). herbivores can alter ecosystems, mostly to the detriment of other species, Most large herbivore species are found in Africa (n = 32), South- including humans, through the loss of ecological interactions and eco- east Asia (n =19),India(n = 14), China (n =14),andtherestofAsia system services. We end by outlining future directions for research (n = 19) (Fig. 1A). Fewer species are found in Europe (n =7),Latin America (n = 5), and North America (n = 5) (fig. S1). Overall, 71 species occur in developing countries, whereas only 10 occur in devel- 1Trophic Cascades Program, Department of Forest Ecosystems and Society, Oregon State oped countries. The highest number of threatened large herbivores University, Corvallis, OR 97331, USA. 2Desert Ecology Research Group, School of Biological occurs in Southeast Asia (n = 19, east of India and south of China), Sciences, The University of Sydney, Sydney, New South Wales 2006, Australia. 3Department of Biology, Stanford University, Stanford, CA 94305, USA. 4Centre for African followed by Africa (n = 12), India (n =9),China(n =8),LatinAmer- Conservation Ecology, Department of Zoology, Nelson Mandela Metropolitan University, ica (n =4),andEurope(n =1)(Fig.1Bandfig.S1).Notably,allofthe Port Elizabeth 6031, South Africa. 5Departamento de Ecologia, Universidade Estadual threatened species of large herbivores are found in developing coun- 6 Paulista (UNESP), C.P. 199, Rio Claro, São Paulo 13506-900, Brazil. College of Natural tries, with the exception of European bison (Bison bonasus), with de- Sciences, Bangor University, Thoday Building, Deiniol Road, Bangor, Gwynedd LL572UW, UK. 7Department of Fisheries and Wildlife, Oregon State University, Corvallis, OR 97331, veloped countries having already lost most of their large mammals in USA. 8Lion Program, Panthera, 8 West 40th Street, 18th Floor, New York, NY 10018, USA. the ongoing megafauna extinction (2). 9Mammal Research Institute, Department of Zoology and Entomology, University of Ecoregions [n = 30, based on (3)] with the most-threatened large Pretoria, Pretoria, Gauteng 0001, South Africa. 10Wildlife Conservation Research Unit, ≥ Department of Zoology, University of Oxford, Recanati-Kaplan Centre, Tubney House, herbivore species ( 5) are found in southern Asia, throughout much Tubney, Abingdon OX13 5QL, UK. 11Environmental Change Institute, School of of extreme Southeast Asia, as well as Ethiopia and Somalia of eastern Geography and the Environment, University of Oxford, Oxford OX1 3QY, UK. 12Nicholas Africa (Fig. 1B and tables S2 to S4). The ecoregions with seven threatened School of the Environment and Earth Sciences, Duke University, P. O. Box 90381, Durham, large herbivore species are the Himalayan subtropical broadleaf forests, NC 27708, USA. 13Department of Ecology and Evolutionary Biology, University of California, Los Angeles, Los Angeles, CA 90095–7239, USA. the Sunda Shelf mangroves, and the peninsular Malaysian rain forests *Corresponding author. E-mail: [email protected] (table S4). Hunting for meat is the predominant threat in all ecoregions
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 1of12 REVIEW
including four critically endangered, and the white rhinoceros is nearly threatened with the current poaching crisis likely to alter its status downward in the near future. Ironically, this endangerment follows one of the greatest success stories in the histo- ry of modern conservation: the recovery of the southern white rhino (C. simum simum) from a single population of fewer than 100 individuals in the early 1900s to about 20,000 today [(7), table S1]. Even with the current crisis of rhinoceros poach- ing, this illustrates that, with sufficient protection, recovery is possible for rela- tively slow-breeding species that are highly prized by poachers. Many of the largest herbivore species have ranges that are collapsing (8, 9). Es- timates of range contractions have been made for 25 of the 74 species, and on av- erage, these species currently occupy only 19% of their historical ranges (table S1). This is exemplified by the elephant, hippo- potamus, and black rhinoceros, all of which now occupy just tiny fractions of their his- torical ranges in Africa (Fig. 3). Further- more, many of these declining species are poorly known scientifically, and badly in need of basic ecological research. Scientif- ic research effort, as measured by the num- ber of published articles on each species, has been much greater for nonthreatened (x̄= 296, SĒx = 129) than threatened spe- cies (x̄= 100, SĒx = 33), and greater over- all for species in developed countries (x̄= Fig. 1. Large herbivore total species richness (A) and threatened (B) at the ecoregion level. 790, SĒx = 376) than developing coun- Ecoregion lists for each species were obtained using the IUCN Red List species range maps and (3) ̄ and are based on the ecoregions where each species is native and currently present. tries (x = 172, SĒx = 33). Indeed, those that have been most studied are primarily game species in wealthy countries, includ- containing at least five threatened large herbivore species (table S2). ing red deer (Cervus elephus), reindeer (Rangifer tarandus), and moose/ These ecoregions fall mostly within the tropical and subtropical moist Eurasian elk (A. alces) (fig. S2). In contrast, 18 of the large herbivore spe- broadleaf forests biome (20 of 30 ecoregions), but biomes containing cies from developing regions have been featured in fewer than 10 pub- combinations of grasslands, shrublands, savannas, mangroves, or other lished articles each (fig. S2), which, in part, reflects negative or indifferent forest types represent the other 10 ecoregions with at least five threat- attitudes toward some species, or low levels of scientific funding, making ened large herbivore species (table S3). it difficult to garner government and public support for scientific studies All 10 large herbivore species within the families Elephantidae, and conservation of these taxa (10). For example, although highly threat- Hippopotamidae, Hominidae, and Tapiridae are currently threatened ened, the six large-bodied species in the Suidae family are collectively (Fig. 2). The large herbivores of the families Suidae, Rhinocerotidae, represented by only 26 published articles (x̄= 4 per species, range = Equidae, and Camelidae are also highly endangered, with 15 of 20 mem- 0 to 14) (table S5 and fig. S2). ber species threatened (Fig. 2). Only eight terrestrial megafauna spe- Between 1996 and 2008, the conservation status of seven herbivore cies (≥1000 kg) exist today as opposed to more than five times that species ≥100 kg deteriorated, whereas only two species improved number (~42) that were present in the late Pleistocene (4–6). The eight (table S1). By contrast, small herbivores are doing relatively well with just remaining species are split between Africa (African elephant, Loxodonta 16% of species below 5 kg in body mass classified as threatened (fig. S3). africana, hippopotamus, Hippopotamus amphibius,andthewhiteand In contrast to the developing world, effective game laws and extirpation of black rhinoceros, Ceratotherium simum and Diceros bicornis,respec- large predators in developed countries of northern latitudes have frequently tively) and Southeast Asia (Asian elephant, Elephas maximus,andthe resulted in an overabundance of large herbivores. In the absence of wolves Indian, Javan, and Sumatran rhinoceros, Rhinoceros unicornis, R. sondaicus, (Canis lupus) and other large carnivores, overabundant cervids can neg- and Dicerorhinus sumatrensis,respectively).Ofthese,sevenarethreatened, atively impact biodiversity, stream morphology, carbon sequestration,
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 2of12 REVIEW
high in tropical forests, leading to vertebrate extirpations on large spatial scales, a process originally dubbed the “empty forest” syndrome (21). Annual consumption of wildlife meat was estimated to be 23,500 tons in Sarawak, Malaysia (22), and 89,000 tons in the Brazilian Amazon (23). Wild meat hunting also represents an increasing threat in Afri- can savannas, resulting in widespread declines in herbivore popula- tions (17). Because wildlife populations outside of protected areas wane, hunters are shifting their attention more to populations in protected areas (17). Demand for wild meat is intensifying, supply is declining, and protected area management budgets for protecting wildlife from over- hunting are often inadequate, particularly in developing nations. This creates a “perfect storm,” whereby overhunting often imparts catastroph- ic population declines (17). Between 1970 and 2005, large mammal populations in Africa’s protected areas decreased by about 59% (16). In part due to overhunting, current ungulate biomass was recently calcu- lated to be only 21% of estimated potential ungulate biomass in Zambia’s national parks (24). Hunting large herbivores for body parts is also driving down pop- ulations of some species, especially the iconic ones. Organized crime is Fig. 2. Proportion of large herbivore species listed as threatened by facilitating a dramatic decline of elephants and rhinoceros in parts of IUCN. The total number of herbivore species in each family is shown after Africa and southern Asia, reversing decades of conservation accom- each family name. Individual threatened species by family include Elephan- plishments. Poaching and illegal trade in elephant products are cur- tidae: African elephant (VU), Asian elephant (EN); Hippopotamidae: hippo- rently the top threats to elephants (25). Ivory poaching has surged potamus (VU), pygmy hippopotamus (EN); Hominidae: eastern gorilla (EN), in recent years, largely due to a rise in demand for and price of ivory western gorilla (EN); Tapiridae: Malayan tapir (VU), Baird’s tapir (EN), lowland in China (26). The number of forest elephants (L. africana cyclotis)in tapir (VU), mountain tapir (EN); Suidae: Philippine warty pig (VU), Oliver’s central Africa declined by 62% between 2002 and 2011 (25). Current- warty pig (EN), Visayan warty pig (CR), Palawan bearded pig (VU), bearded ly, 75% of elephant populations are declining and at risk of extirpa- pig (VU); Rhinocerotidae: Indian rhinoceros (CR), Javan rhinoceros (CR), tion, and the range of elephants has drastically declined (26). More Sumatran rhinoceros (CR), black rhinoceros (CR); Equidae: Grevy’s zebra than 100,000 African elephants were poached during the 3-year period (EN), mountain zebra (VU), African wild ass (CR), Przewalski’s horse (EN), Asiatic wild ass (CR); Cervidae: sambar (VU), barasingha (VU), Père David’s from 2010 to 2012 (26). This level of illegal kills represents 20% of the deer (EW), white-lipped deer (VU); Camelidae: bactrian camel (CR); Bovidae: current estimated population size of 500,000 African elephants, and Indian water buffalo (EN), gaur (VU), kouprey (CR), European bison (VU), even populations of savanna (or bush) elephants (L. africana africana) wild yak (VU), banteng (EN), takin (VU), lowland anoa (EN), tamaraw (CR), are now declining (26). Poaching of rhinoceros for their horns has also mountain nyala (EN), scimitar-horned oryx (EW), mountain anoa (EN), soared in recent years because of its use in traditional Chinese med- Sumatran serow (VU), walia ibex (EN). Scientific names in table S1. icine. The number of rhinoceros poached in South Africa grew by two orders of magnitude from 13 in 2007 to 668 in 2012 (27) and 1004 in 2013 (28). The situation is so desperate that an emergency interven- and ecosystem function (11). Confining large herbivores within fixed tion is planned in which large numbers of white rhinoceros will be boundaries can also lead to overabundance as with bison (Bison bison) translocated out of South Africa’sKrugerNationalParkandplaced in North America (12) and elephants in Africa (13). in potentially more secure areas (29). Furthermore, at least in part due to poaching, Africa’s western black rhinoceros (D. bicornis longipes) was declared extinct in 2011 (1). This slaughter is driven by the high THREATS retail price of rhinoceros horn, which exceeds, per unit weight, that of gold, diamonds, or cocaine (27). If accelerated poaching by organized The main threats to large herbivores are hunting, competition with live- crime syndicates continues, Africa’s rhinoceroses may become extinct stock, and land-use change such as habitat loss, human encroachment, in the wild within 20 years (27). Numerous species of other large her- cultivation, and deforestation (Fig. 4 and fig. S4). Extensive overhunting bivores are also hunted for their body parts, including hippopotamus formeatacrossmuchofthedevelopingworldislikelythemostimpor- fortheirivoryteeth,bovidsforhornsandskulls,equidsforhides,ta- tant factor in the decline of the largest terrestrial herbivores (14–17). pirs for feet and hides, cervids for antlers, giraffids for hides, and goril- Slow reproduction makes large herbivores particularly vulnerable to las for heads, hands, and feet (1). Large herbivores are more vulnerable overhunting. The largest- and slowest-to-reproduce species typically than smaller herbivores to overharvesting through a combination of the vanish first, and as they disappear, hunters turn to smaller and more generally higher value of larger bodies or their parts, and the slow life fecund species (14), a cascading process that has likely been repeated history of the larger herbivores. Together, these increase the likelihood for thousands of years (6, 18, 19). In synergy with changes in land use, of large herbivores being harvested and reduce their ability to recover hunting for meat has increased in recent years due to human popu- from such harvests. lation growth, greater access to wildlands due to road building, use of Livestock continues to encroach on land needed for wild grazers modern firearms and wire snares, access to markets, and the rising de- and browsers, particularly in developing countries where livestock pro- mand for wild meat (14, 20). Wild meat harvests have been especially duction tripled between 1980 and 2002 (30). There are an estimated 3.6
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 3of12 REVIEW
Fig. 3. Range contractions over time for three iconic African herbi- and distances. The black rhinoceros range has continued to shrink since vores. African elephant (ca. 1600 versus 2008), common hippopotamus 1987 across most of Africa, but has expanded locally in Zambia, South (ca. 1959 versus 2008), and black rhinoceros (ca. 1700 versus 1987). The his- Africa, and Namibia through recent reintroductions, and the most current torical ranges are in blue, whereas the most recent ranges are represented range polygons are not shown because of the recent poaching pressure by darker-colored polygons. For security purposes, the most recent black on the rhinoceros. Photo Credits: Elephant and hippopotamus (K. Everatt), rhinoceros range polygons (1987) have been moved by random directions rhinoceros (G. Kerley).
billion ruminant livestock on Earth today, and about 25 million have been added to the planet every year (~2 million/month) for the last 50 years (31). This upsurge in livestock has resulted in more competition for grazing, a reduction in forage and water available to wild herbi- vores, a greater risk of disease transmission from domestic to wild spe- cies (32), and increased methane emissions (31). In central Asia, the expansion of goat grazing for cashmere wool production for interna- tional export has reduced habitats available to large herbivores with con- sequent impacts on their predators including snow leopards (Panthera uncia)(33). Livestock competition is also a significant threat to large her- bivores elsewhere in Asia, with multiple species jeopardized by this threat in India (n =7),China(n =7),andMongolia(n =4)(fig.S4).Hybrid- ization with domestic livestock varieties is also a serious problem for some wild species such as the Indian water buffalo (Bubalus arnee), Bactrian camel (Camelus ferus), wild yak (Bos mutus), Przewalski’s horse (Equus ferus), and several wild pig species (Sus spp.) in South- east Asia (1). Ironically, in many pastoral settings in Africa, domestic livestock are abundant but not regularly consumed for subsistence, and are instead kept as a means of storing wealth, as a status symbol, or for consumption on special occasions (14). Livestock is a private good, and so, people invest significant energy to protect it, whereas wild her- bivores are typically a public good, often resulting in weak incentives for their conservation and in many cases open access to the resource, Fig. 4. Proximate threats faced by large herbivores globally. Threats both of which commonly result in overuse. faced by each species were categorized using information in the IUCN Habitatlossisasignificantthreattolargeherbivoresinpartsof Red List species fact sheets. The total adds up to more than 100% because Latin America, Africa, and Southeast Asia (Fig. 4). The causes of this each large herbivore species may have more than one existing threat. threat have important drivers originating in developed countries due
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 4of12 REVIEW
Fig. 5. Conceptual diagrams showing the effects of elephants, hippopotamus, and rhinoceros on ecosystems. (A) African elephants (L. africana)con- vert woodland to shrubland (53), which indirectly im- proves the browse availability for impala (A. melampus) (53) and black rhinoceros (D. bicornis minor)(54). By damaging trees, African elephants facilitate increased structural habitat complexity benefiting lizard com- munities (100). Predation by large predators (for exam- ple, lions) on small ungulates is facilitated when African elephants open impenetrable thickets (48). African elephants are also great dispersers of seeds over long distances (13). (B) Hippopotamus (H. amphibius) maintain pathways in swamps, leading to new chan- nel systems (101). Areas grazed by hippopotamus are often more nutritious, which benefits kob (K. kob) (55). Mutualism and semiparasitism between hippo- potamus and birds have also been shown, via the lat- ter eating insects on hippopotamus (73). (C)White rhinoceros (C. simum) maintain short grass patches in mesic areas, which increases browse for other grazers (impalas, wildebeests, C. taurinus, and zebra, Equus burchelli) and changes fire regimes (71). to demand for agricultural and other products. Southeast Asia has the highest rate of deforesta- tion among tropical regions, and if trends contin- ue, Southeast Asia could lose 75% of its original forests and nearly half of its biodiversity by the end of this century (34). Habitat loss is typically asymmetrical with respect to quality, with remain- ing habitat generally being less productive. A simi- lar trend is found in the tendency to create protected areas in steep, rocky, or dry terrain (35), trapping species of conservation concern in suboptimal ha- bitats (36). Additionally, the greater area requirements of larger species make them unable to persist in smaller fragments of habitat, which may still sup- port smaller herbivores. Their larger area require- ment also makes larger species that persist in fragments increasingly susceptible to conservation challenges that affect small populations. This sug- gests a greater likelihood of extinction among the larger rather than smaller herbivores. Other threats to large herbivores include human encroachment (including road building), cultiva- tion of crops, and civil unrest, all of which contrib- ute to population decline (Fig. 4). In the future, synergies among the factors discussed here will ex- acerbate the dangers to large herbivores, as is the case when increased hunting results from people being given access to fragmented, isolated forest remnants within previously extensive and less ac- cessible areas (19). Beyond declines in abundance, the most threatened large herbivores are further imperiled by a loss of genetic diversity. The European bison, for instance, passed through a severe genetic bottleneck in the early 20th century and now suffers from balanoposthitis, a necrotic inflammation of the prepuce that inhibits breeding (37).
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 5of12 REVIEW
CONSEQUENCES OF LARGE HERBIVORE DECLINE Synergy between herbivores Megaherbivores, primarily via their effects on vegetation structure, can Large herbivores shape the structure and function of landscapes and facilitate the existence and survival of a suite of mesoherbivores. For environments in which they occur (Fig. 5). They directly and in- example, in northern Botswana, browsing by African elephants helps directly affect other animal species throughout the food web, including convert woodland to shrubland, increasing the dry season browse for their predators and smaller herbivores, and modify abiotic processes impalas (Aepyceros melampus)(53). In Addo Elephant National Park, involving nutrient cycles, soil properties, fire regimes, and primary South Africa, African elephants create pathways in impenetrable thickets, production. The roles of large herbivores thus cannot be taken over facilitating black rhino browsing (54). In some seasons, areas grazed by or compensated for by smaller herbivores. These effects of large her- hippopotamus in Benue National Park, Cameroon, are more nutritious bivores on ecosystems are further discussed below. with regard to structure and nutrients, which is advantageous for kob (Kobus kob)(55). In contrast, high densities of large herbivores inside Large herbivores as ecosystem engineers reserves or in the absence of their predators can be detrimental where Large herbivores, through their size and high biomass, exert many di- overgrazing decreases foraging opportunities for coexisting browsers (56), rect effects on vegetation via trampling and consumption of plants particularly during periods of low rainfall (57). However, by generally pro- (38). Hence, they maintain patch heterogeneity in systems that would moting the replacement of tall mature woodlands or grasslands by ra- otherwise support continuous woody vegetation. Even in wetter cli- pidly growing shrubs or short grasses, large herbivores are more likely mates, which favor trees over grasses, elephants can maintain open to have positive than negative impacts on mesoherbivores (38). patches (39). Bison also maintain and expand grasslands, and their wal- lows increase habitat diversity for a variety of both plants and animals Seed dispersal (40). Indeed, the larger herbivores consume and, hence, influence the Extinct megaherbivores once played a critical role in the colonization fate of a larger variety of plant species than coexisting mesoherbivores of woody plants (58). Even today, large herbivores are irreplaceable as (13). The Pleistocene megafauna extinction can be viewed as a global-scale seed dispersers because, relative to smaller frugivores, they are able to natural experiment that highlights the continental scale of the ecological consume larger seeds and deliver many more seeds per defecation event impacts that result from the loss of large herbivores. Evidence from over longer distances. Elephants may consume more seeds from a greater Australia suggests that mixed rainforest was converted to sclerophyll number of species than any other taxon of large vertebrate (13, 59, 60). vegetation in the aftermath of megafaunal loss (41), whereas in North In Congo alone, forest elephants (L. africana cyclotis) disperse ca. 345 America, novel plant communities formed that have no modern analogs large seeds per day from 96 species, consistently more than 1 km from (42), and in Europe, a heterogeneous mosaic of vegetation structures was the parent trees (61). Indian rhinoceros (R. unicornis)movelargetree replaced with more closed woodland communities (43) as a result of seeds from forest canopies to grasslands, generally with successful ger- the particularly severe megafaunal declines in these regions (2). mination and recruitment (62). Even smaller species, such as tapirs (Tapirus spp.) and gorillas (Gorilla gorilla) are effective seed dispersers, Predators and scavengers which helps to maintain the distribution and abundance of plant species Large herbivores are the primary source of food for predators and (63, 64). For instance, in African lowland rainforests, primate-dispersed scavengers that have high energetic demands, making them an integral tree species were less abundant at sites with depleted primate popula- component of the food web (11). Lions (Panthera leo) and spotted tions due to intense hunting by humans compared with sites with low hyenas (Crocuta crocuta) prefer prey above ~90-kg body mass, and hunting pressure (65). Thus, the loss of large seed dispersers may lead all of the world’s largest terrestrial carnivores prey on large herbivores to a wave of recruitment failures among animal-dispersed species (66) (11, 44). Indeed, even the megaherbivores (≥1000kg)suchasele- with potential consequences for important ecological services (67). phants are not immune to predation (45), because their juveniles are within the size range preferred by some large carnivores (46). Notably, Nutrient cycling large herbivores may even facilitate the hunting success of predators Large herbivore communities consume disproportionately more plant when their foraging activities open up dense vegetation, making small biomass per unit area than small herbivores (68). They affect nutrient herbivores more vulnerable (47, 48). Large herbivore carcasses yield cycles via direct and indirect mechanisms that have consequences for more nutrients to a wider suite of scavengers than those of smaller ecosystem functioning. For example, large herbivores directly influ- species because the latter are usually consumed completely, whereas ence nutrient cycling via the consumption of plants, which indirectly large carnivores tend to consume relatively less of large carcasses, there- causes the reallocation of carbon and nutrients within the plant, while by leaving more for other species (49). In Yellowstone National Park, also shifting plant species composition toward species with different gray wolves have been shown to buffer the negative impacts of shorter ratesoflitterdecomposition(69). Herbivores can greatly accelerate winters through the food subsidies they provide for a suite of scavengers the nutrient cycle in ecosystems through consumption and subsequent [for example, coyotes (Canis latrans), foxes (Vulpes vulpes), ravens defecation, returning nutrients to the soil at rates that are orders of (Corvus corax), and eagles (Haliaeetus spp.)] (50). Given the pivotal magnitude faster than processes of leaf loss and decay. Moreover, as and positive role of top predators in many ecosystems, it is unfortu- leaves and twigs are consumed, largeherbivoresexcreteurineandfe- nate that depletion of their prey is a serious threat in developing coun- ces and create patches of concentrated nutrients that can last for sev- tries (11, 33, 51), particularly for obligatemeateaterssuchasjaguars eral years (69). On longer time scales, as the location of concentrated (Panthera onca), tigers (Panthera tigris), lions, leopards (Panthera pardus), patchesshiftsovertime,largeherbivoresmayplayadisproportionate and snow leopards (P. uncia). For example, overhunting of large herbi- role in diffusing nutrients across landscapes (68). Carcasses also add a vores in West Africa has reduced the prey base, which, at least in part, variety of nutrients to the soil such as calcium, with effects that can has caused regional lion populations to become critically endangered (52). persist several years after the death of the animal (68, 70).
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 6of12 REVIEW
Fire resources essential for ensuring the preservation of our global natural By altering the quantity and distribution of fuel supplies, large herbivores heritage of large herbivores. A sense of justice and development is es- can shape the frequency, intensity, and spatial distribution of fires across sential to ensure that local populations can benefit fairly from large a landscape. There are even unique interactions among large herbivore herbivore protection and thereby have a vested interest in it. The pres- populations that can influence fire regimes. For example, facilitative inter- ence of a diversity of large charismatic species can yield financial actions between white rhinoceros and mesoherbivores result in reduced benefits that flow to local communities (80). For example, with the Af- fuel loads and fuel continuity, and consequently fewer large, intense fires rican photo safari industry, the prospect of simply observing large car- (71). Other factors can influence the frequency and intensity of fires, partic- nivores, elephants, or rhinoceros can drive tourism revenue. The ularly in locations where the total area burned is strongly related to ungulate ultimate forces behind declining large mammal populations are a rising population size. For example, Serengeti wildebeest (Connochaetes taurinus) human population and increasing per capita resource consumption populations irrupted after the rinderpest virus was eradicated in the 1960s, (Fig. 7). As is the case for the conservation of most taxa, programs and the subsequent increase in grazing pressure led to a widespread reduc- tion in the extent of fires and delayed recovery of tree populations (72). The removal of plant biomass by browsing also reduces fire fuel loads and decreases fire susceptibility. Thus, there is scant evidence of fire in much of Australia until the megafauna disappeared after humans arrived (5).
Small animals Despite huge differences in body size, large herbivores interact with a suite of small animals including birds, insects, rodents, lizards, and others (Fig. 5). For example, several fish species feed on flesh wounds of hippopotamus (73), and the dung of Asian elephants may be used by amphibians as daytime refuge, particularly in the dry season when leaf litter is scarce (74). Bison wallows support amphibians and birds by creating ephemeral pools, and bison grazing may facilitate habitat for prairie dogs (Cynomys spp.) and pocket gophers (geomyids) (40). Oxpeckers (Buphagus spp.) depend on the large herbivores for their diet of ectoparasites, and blood-sucking insects such as tsetse flies (Glossina spp.) largely depend on herbivores for food. The presence of large herbivores can also reduce the negative effects of rodent out- breaks. For example, in Kenya, the pouched mouse (Saccostomus mearnsi) markedly increased in density after the exclusion of large herbivores, due to an increase in the availability and quality of food (75). Thus, a Fig. 6. Photos of selected threatened large herbivore species. Endan- reduction in large herbivore populations could have unintended conse- germent status and photo credits include the following: lowland tapir quences if rodent abundance increases, particularly if there are (i) neg- (Tapirus terrestris), vulnerable, T. Newsome; mountain nyala (T. buxtoni), endangered, H. Hrabar; European bison (B. bonasus), vulnerable, G. Kerley; ative effects on plant communities, (ii) increased risks of rodent-borne eastern gorilla (Gorilla beringei), endangered, P. Stoel; mountain zebra diseases, or (iii) increases in predators that specialize on rodents (76, 77). (Equus zebra), vulnerable, H. Hrabar. Humans The loss of large herbivores has direct effects on humans, especially for food security in developing regions. It is estimated that 1 billion people rely on wild meat for subsistence (15). Under a business-as-usual scenario, food security will continue to falter given that wild meat in African forests is expected to decline by more than 80% during the next 50 years (78). Moreover, charismatic large herbivores are important flagship fauna (Fig. 6) that draw many tourists to protected areas, espe- cially when they are sympatric with large carnivores (79). Although the consistency of ecotourism can be interrupted by unpredictable events such as disease epidemics and civil unrest, a decline of large flagship species translates directly into reduced tourism (animal watching, photo and hunting safaris) and thereby a decline in trade balances and em- ployment, particularly in rural parts of the developing world where most megaherbivores persist and poverty is common. Fig. 7. Global change in the collective mass for wild mammals, hu- FUTURE DIRECTIONS mans, cattle, and all livestock for the years 1900–2050. Values for 1900 and 2000 are from (102). Human, cattle, and livestock biomass Saving the remaining threatened large herbivores will require concerted forecasts are based on projected annual growth in human population, beef action. The world’s wealthier populations will need to provide the production, and meat production, respectively (88, 102).
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 7of12 REVIEW that help to lower human birth rates in rapidly growing regions such Addressing poaching as those that enhance educational and development opportunities, Solving the current crisis associated with poaching for meat and body particularly for young women, are a high priority. However, the reality parts is an essential step, although one that is extremely challenging. is that strategies for conserving herbivores in the context of high human Trade bans alone can sometimes succeed but can also fail because they population densities are likely to be increasingly important. Increasing limit supply, causing prices to rise, thereby driving more poaching for levels of human carnivory are at the crux of the problem. Lowering hu- the black market (27). Multifaceted bold new policies are urgently man consumption of domestic ruminants could help conserve herbi- needed that (i) increase the effectiveness of law enforcement both through vore populations by reducing demand for rangeland forage, water, antipoaching and strengthened penal systems related to poaching, (ii) and feed crops. Reducing consumption of wild herbivores can also be incentivize local communities to conserve wildlife (for example, in- effective, and enforced wildlife management such as via wildlife ranch- creasing tourism income), (iii) reduce demand for illegally sourced ing has proven to be very successful at maintaining sustainably high har- wildlife products through market mechanisms of controlled trade of vests of wild meat while providing subsistence food resources to local products or farming animals (17, 81), and (iv) aid a cultural shift away people. The implementation of wildlife management strategies such as from luxury wildlife products in industrializing countries such as China male-only harvests, age-specific harvests, and quotas has the potential and Vietnam. Social marketing and environmental education programs to improve both conservation and food security if improved governance can also be highly effective in reducing demand for wildlife. For ex- can allow for implementation of these strategies. In the near future, ample, shark fin sales plummeted after social media pleas by basketball urgent action is needed to prevent the extinction of species with ex- celebrity Yao Ming. Likewise, other prominent Chinese celebrities tremely low populations, especially those with limited captive popula- have also started speaking out to reduce demand for ivory and rhinoceros tions(forexample,Bactriancamel,rhinos, and suids). Decisive steps horn in Southeast Asia. will be required to address key threats facing threatened large herbi- vores including, among others, the following. Managing protected areas Globally, only ~10% of conservation funding for protected areas is Focusing research efforts spent in developing countries (82). Underfunding of protected area Basic data and information on the status and ecology of a significant networks, particularly in the tropics, results in failure to control key number of large herbivore species are still lacking. From a conserva- threats to herbivores. In the absence of funds for law enforcement, tion perspective, we call for a major shift in the large herbivore re- poaching for meat or body parts proceeds unhindered, and many pro- search effort from the few nonthreatened species in developed countries tected areas are being encroached by human settlement, livestock, and (for example, red deer, reindeer, and moose/Eurasian elk) to the many logging. Large herbivores, including those that are migratory, need threatened species in developing countries (43 species; fig. S2). We ur- large areas to support viable populations. Given the global tendency gently recommend more research on the most threatened large herbi- for protected areas to be small (<10,000 ha), many protected areas are vores in Southeast Asia, Africa, and Latin America. Species in need of unable to effectively contribute to the persistence of large herbivores immediate attention include the critically endangered tamaraw (Bubalus (83). Therefore, expanding protected areas and increasing connectivity mindorensis), Visayan warty pig (Sus cebifrons), and walia ibex (Capra between them are important. In some contexts, fencing can assist by walie), as well as the endangered Oliver’swartypig(Sus oliveri), moun- demarcating boundaries and reducing human encroachment, while at tain anoa (Tragelaphus buxtoni), lowland anoa (Bubalus depressicornis), the same time reducing edge effects and making law enforcement and mountain tapir (Tapirus pinchaque), all having fewer than 10 pub- easier (84). Technological approaches such as the use of drones may lished articles per species (fig. S2). In particular, more research is needed help to patrol parks with limited resources, but for effectiveness, this to understand the various ways that rising human and livestock den- technology will need to be low cost, easy to use, durable, and efficient. sities (Fig. 7), changing climate, habitat loss, and hunting, as well as Without the cooperation of people who live near wildlife, conservation different combinations of these factors, affect these large herbivores. efforts are likely to fail. To ensure just outcomes, it is essential that We urge large carnivore researchers and conservation agencies to in- local people be involved in and benefit from the management of vest more money and attention on the large herbivores that comprise protected areas. Local community participation in the management large carnivore prey, because depletion of prey is a significant global of protected areas is highly correlated with protected area policy com- threat to large carnivores (11). In an attempt to shift the research effort pliance (85). For instance, to protect wildlife, Nepal has successfully from well-studied species in developed countries to highly threatened adopted a policy of sharing of revenues from protected areas with local species in developing countries, we recommend the establishment people who live adjacent to the reserves (86). of a fund to finance graduate students to conduct empirical ecolog- ical and socioeconomic research that would benefit endangered Focusing conservation efforts large herbivores. Examples of potential thesis topics could include In southern Asia and other developing areas, oil palm plantations, (i) replicated studies of the basic ecology of large, rare herbivores that pulp and paper, and other commodity crops are rapidly replacing are the least studied, (ii) seed dispersal and woody flora recruitment in wet tropical forests where large herbivore populations are at risk. In areas with and without large herbivores, (iii) effects of diversity of this situation, it makes sense to shift agricultural expansion to abun- large herbivore species on financial benefits flowing to communities dant degraded low-carbon density lands while sparing the high carbon from tourism, (iv) success of stall-feeding livestock programs for stock lands for climate change mitigation and animal conservation potentially reducing competition between livestock and wild herbi- (87). Infrastructure and mining development are additional important vores, and (v) potential for increases in traditionally grown protein- factors in habitat loss. Initiatives are needed to encourage mining rich plant foods rather than domestic or wild meat as a primary companies to underwrite conservation efforts. Moreover, mining sites protein source for humans. couldbeusedasdefactowildliferefugesbybringingsecuritytoplaces
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 8of12 REVIEW that lack it, in turn providing a safe haven for large mammals away species may be slipping away even before they are discovered and described from poachers. Moreover, the move of poor rural populations to cities by science. Recently, two rare large herbivores were discovered: a fifth spe- and towns leaves a great opportunity for restoration of large mammals cies of tapir, the kabomani tapir (Tapirus kabomani), in the Amazon (92) in the hinterlands. and a bovid, the saola (Pseudoryx nghetinhensis), in Southeast Asia (93). Africa has more large herbivores than any other world region and To jump-start protection for the saola, conservationists recently removed lower endangerment rates (12 of 32 are threatened) than any other 27,000 snares from the forests of Vietnam and Laos (94). region in the developing world (fig. S1). However, over the next half The problem of large herbivore declines may not be solved by the century, sub-Saharan Africa will have the world’shighestprojected current Convention on Biological Diversity target of protecting 17% of growth rates of human population and livestock production (88) terrestrial land by 2020 (95). Given the substantial area requirements (fig. S5), which are potential drivers of hunting for meat, habitat loss, of large herbivores, 17% of land in isolated fragments is unlikely to pro- and livestock competition. With land use and human demographic vide sufficient protection to slow or reverse declines, particularly given patterns in sub-Saharan Africa becoming more similar to those in that inadequate policing/funding can effectively reduce the size of Southeast Asia, where all 19 large herbivore species are threatened with protected areas (96). This is further exacerbated by the global tendency extinction, it is critical to develop a geographic approach to conservation for protected areas to be in low-quality habitats (35), effectively reducing the that focuses on areas with high species diversity, and this should address densities, and hence numbers, that can be conserved in these areas (36). both human issues (as indicated above) and conservation management. The range contractions (fig. S6) and population declines of large Additionally, conservation actions are dependent on available money. herbivore species have ecological and evolutionary implications. Range We advocate for a global government-funded scheme for rare large her- contractions inevitably result from the loss of local populations, many bivores beyond elephants and rhinoceros, as well as the establishment of which are genetically distinct, thus representing a major and under- of a nongovernmental organization that focuses exclusively on rare appreciated pulse of biological extinction (97). Even if they survive in large herbivores, like what the Arcus Foundation does for apes or what protected areas, many of these largest species might already be below Panthera does for large cats. the minimum numbers to be effective in generating ecological cas- cades (Fig. 5) or allowing evolutionary processes such as speciation Addressing climate change (98). Furthermore, 11 of the 44 threatened species are on the Evolu- There are potential combined strategies (joined-up policies) that would tionarily Distinct and Globally Endangered (EDGE) list due to their mitigateclimatechangeandatthesametimebenefitlargeherbivores. unique characteristics while being on the verge of extinction. These are Examples include (i) curbing ruminant livestock numbers while in- the mountain, Asian, and Baird’stapirs(Tapirus spp.), black, Javan, creasing high-protein plant-based foods or nonruminant meat so as to and Sumatran rhinoceros, Bactrian camel, Asian elephant (E. maximus), lower greenhouse gas (for example, methane and carbon) emissions pigmy hippopotamus (Choeropsis liberiensis), African wild ass (Equus while also reducing competition with large herbivores (31), and (ii) en- africanus), and western gorilla (99). Thousands of years ago, equids hancing carbon storage by preventing tropical forests from being logged were among the most abundant large grazing animals of the grasslands while also protecting habitat for large herbivores. In addition, tropical and steppes of Africa, Asia, and the Americas, whereas today, after large herbivores disperse large seeds that are typically from slow-growing many of their populations have been decimated, five of the remaining and densely growing tree species important for carbon storage. By 2050, seven species are threatened and at risk of extinction (1). These are the climate change has the potential to leave many of Earth’s species destined African wild ass, Asiatic wild ass (Equus hemionus), Przewalski’s horse, for extinction (89). Additional research is urgently needed to better Grevy’szebra(Equus grevyi), and mountain zebra (E. zebra). predict changes in large herbivore population sizes and ranges with Growing human populations, unsustainable hunting, high densities climate change while accounting for the current threats they face. of livestock, and habitat loss have devastating consequences for large, long-lived, slow-breeding, and, therefore, vulnerable herbivore species, their ecosystems, and the services they provide. Large herbivores, and FINAL THOUGHTS their associated ecological functions and services, have already largely been lost from much of the developed world. The scale and rate of The wave of species extinctions that obliterated 80% of the Pleistocene large herbivore decline suggest that without radical intervention, large megaherbivores (≥1000 kg) on planet Earth appears to be continuing herbivores (and many smaller ones) will continue to disappear from today in Africa and Southeast Asia. The very recent extinctions of numerous regions with enormous ecological, social, and economic Africa’s western black rhinoceros and Vietnam’s Javan rhinoceros are costs. We have progressed well beyond the empty forest to early views sober reminders of this long-term trend (1, 90). Then as now, the of the “empty landscape” in desert, grassland, savanna, and forest eco- Pleistocene extinctions were triggered in part by human hunters (2, 91). systems across much of planet Earth. Now is the time to act boldly, be- Solving the current poaching crisis, a sinister development of orga- cause without radical changes in these trends, the extinctions that nized crime, will help but will likely be insufficient to stem, much less eliminated most of the world’s largest herbivores 10,000 to 50,000 years reverse, impending declines and future extinctions among the few re- ago will only have been postponed for these last few remaining giants. maining megafauna. Megafauna remain beset by long-standing and generally escalating threats due to land-use change and ongoing paro- chial poaching by locals. The situation for the 66 species of large her- SUPPLEMENTARY MATERIALS bivores having body masses of 100 to 1000 kg is not as dire as for those ≥ Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/ 1000 kg, but still ominous because 55% of these herbivores are currently full/1/4/e1400103/DC1 threatened (fig. S3). Within this body mass range, hominid, tapirid, suid, Fig. S1. Regional patterns of endangerment of large herbivores. and equid species are the most highly threatened families (Fig. 2). Some Fig. S2. Number of published scientific articles by species.
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 9of12 REVIEW
Fig. S3. Comparison of Pleistocene extinctions by body mass with current threatened species 21. K. H. Redford, The empty forest. BioScience 42, 412–422 (1992). by body mass. 22. E. L. Bennett, Is there a link between wild meat and food security? Conserv. Biol. 16,590–592 Fig. S4. Global distribution of the four main threats faced by large herbivores. (2002). Fig. S5. Human population trends and projections by region (top) and ruminant livestock 23. C. A. Peres, Effects of subsistence hunting on vertebrate community structure in Amazonian trends by region (bottom). forests. Conserv. Biol. 14, 240–253 (2000). Fig. S6. Current range maps (sorted by family) for the 72 large herbivores not classified as 24. P. A. Lindsey, V. R. Nyirenda, J. I. Barnes, M. S. Becker, R. McRobb, C. J. Tambling, W. A. Taylor, extinct in the wild (EW). F. G. Watson, M. t’Sas-Rolfes, Underperformance of African protected area networks Table S1. Data on the 74 large terrestrial herbivores above 100 kg. and the case for new conservation models: Insights from Zambia. PLOS One 9, e94109 (2014). Table S2. The number of large herbivores (threatened, total, and facing each of the four main 25. F. Maisels, S. Strindberg, S. Blake, G. Wittemyer, J. Hart, E. A. Williamson, R. Aba’a, G. Abitsi, threats) found in each ecoregion. R. D. Ambahe, F. Amsini, P. C. Bakabana, T. C. Hicks, R. E. Bayogo, M. Bechem, R. L. Beyers, Table S3. The number of large herbivores (threatened and total) found in each ecoregion. A. N. Bezangoye, P. Boundja, N. Bout, M. E. Akou, L. B. Bene, B. Fosso, E. Greengrass, F. Grossmann, Table S4. The threatened large herbivores found in each of the ecoregions with at least five C. Ikamba-Nkulu, O. Ilambu, B. I. Inogwabini, F. Iyenguet, F. Kiminou, M. Kokangoye, threatened large herbivores. D. Kujirakwinja, S. Latour, I. Liengola, Q. Mackaya, J. Madidi, B. Madzoke, C. Makoumbou, Table S5. Summary of research effort for the period 1965 to June 2014. G. A. Malanda, R. Malonga, O. Mbani, V. A. Mbendzo, E. Ambassa, A. Ekinde, Y. Mihindou, References (103–107) B. J. Morgan, P. Motsaba, G. Moukala, A. Mounguengui, B. S. Mowawa, C. Ndzai, S. Nixon, P. Nkumu, F. Nzolani, L. Pintea, A. Plumptre, H. Rainey, B. B. de Semboli, A. Serckx, E. Stokes, A. Turkalo, H. Vanleeuwe, A. Vosper, Y. Warren, Devastating decline of forest elephants in central Africa. PLOS One 8, e59469 (2013). REFERENCES AND NOTES 26. G. Wittemyer, J. M. Northrup, J. Blanc, I. Douglas-Hamilton, P. Omondi, K. P. Burnham, Illegal killing for ivory drives global decline in African elephants. Proc. Natl. Acad. Sci. 1. IUCN, The IUCN Red List of Threatened Species. Version 2013.2 (2013); http://iucnRedList.org. U.S.A. 111, 13117–13121 (2014). 2. C. Sandom, S. Faurby, B. Sandel, J.-C. Svenning, Global late Quaternary megafauna ex- 27. D. Biggs, F. Courchamp, R. Martin, H. P. Possingham, Legal trade of Africa’s rhino horns. tinctions linked to humans, not climate change. Proc.R.Soc.BBiol.Sci.281, 20133254 Science 339, 1038–1039 (2013). (2014). 28. South African National Parks, Update on Rhino Poaching Statistics (2014); http://sanparks. 3. D. M. Olson, E. Dinerstein, E. D. Wikramanayake, N. D. Burgess, G. V. N. Powell, E. C. Underwood, org/about/news/default.php?id=55976. J. A. D’amico, I. Itoua, H. E. Strand, J. C. Morrison, C. J. Loucks, T. F. Allnutt, T. H. Ricketts, Y. Kura, 29. South African National Parks, Media Release: Environmental Affairs Minister Leads Imple- J.F.Lamoreux,W.W.Wettengel,P.Hedao,K.R.Kassem,Terrestrialecoregionsoftheworld:A mentation of Integrated Strategic Management of Rhinos in SA (2014); http://www.sanparks. new map of life on Earth. BioScience 51, 933–938 (2001). org/about/news/default.php?id=56141. 4. N. Owen-Smith, Pleistocene extinctions: The pivotal role of megaherbivores. Paleobiology 13, 30. World Bank, Minding the Stock: Bringing Public Policy to Bear on Livestock Sector Development. 351–362 (1987). Report No. 44010-GLB (World Bank, Washington, DC, 2009). 5. T. F. Flannery, The Future Eaters: An Ecological History of the Australasian Lands and 31. W. J. Ripple, P. Smith, H. Haberl, S. A. Montzka, C. McAlpine, D. H. Boucher, Ruminants, climate People (Reed Books, Sydney, 1994). change and climate policy. Nat. Clim. Change 4,2–5 (2014). 6. D. K. Grayson, The archeological record of human impacts on animal populations. J. 32. D. P. Mallon, J. Zhigang, Grazers on the plains: Challenges and prospects for large herbivores World Prehistory 15,1–68 (2001). in Central Asia. J. Appl. Ecol. 46,516–519 (2009). 7. R. Emslie, M. Brooks, African Rhino: Status Survey and Conservation Action Plan (IUCN 33. J. Berger, B. Buuveibaatar, C. Mishra, Globalization of the cashmere market and the de- World Conservation Union, Cambridge, 1999). cline of large mammals in central Asia. Conserv. Biol. 27, 679–689 (2013). 8. J. C. Morrison, W. Sechrest, E. Dinerstein, D. S. Wilcove, J. F. Lamoreux, Persistence of 34. N. S. Sodhi, L. P. Koh, B. W. Brook, P. K. L. Ng, Southeast Asian biodiversity: An impending large mammal faunas as indicators of global human impacts. J. Mammal. 88, 1363–1380 disaster. Trends Ecol. Evol. 19, 654–660 (2004). (2007). 35. L. N. Joppa, A. Pfaff, High and far: Biases in the location of protected areas. PLOS One 4, 9. M. Sanjayan, L. H. Samberg, T. Boucher, J. Newby, Intact faunal assemblages in the modern e8273 (2009). era. Conserv. Biol. 26, 724–730 (2012). 36. G. I. H. Kerley, R. Kowalczyk, J. P. G. M. Cromsigt, Conservation implications of the refugee 10. R. A. Blouch, Conservation and research priorities for threatened suids of South and species concept and the European bison: King of the forest or refugee in a marginal Southeast Asia. IBEX JME 3,21–25 (1995). habitat? Ecography 35, 519–529 (2012). 11. W. J. Ripple, J. A. Estes, R. L. Beschta, C. C. Wilmers, E. G. Ritchie, M. Hebblewhite, J. Berger, 37. M. W. Hayward, R. Kowalczyk, Z. A. Krasiński, M. Krasińska, J. Dackiewicz, T. Cornulier, B.Elmhagen,M.Letnic,M.P.Nelson,O.J.Schmitz,D.W.Smith,A.D.Wallach,A.J.Wirsing, Restoration and intensive management have no effect on evolutionary strategies. Endanger. Status and ecological effects of the world’s largest carnivores. Science 343, 1241484 Species Res. 15,53–61 (2011). (2014). 38. N. Owen-Smith, Megaherbivores: The Influence of Very Large Body Size on Ecology (Cambridge 12. L. E. Painter, W. J. Ripple, Effects of bison on willow and cottonwood in northern Univ. Press, Cambridge, 1988). Yellowstone National Park. Forest Ecol. Manag. 264, 150–158 (2012). 39. R. T. Corlett, The shifted baseline: Prehistoric defaunation in the tropics and its 13. G. I. H. Kerley, M. Landman, The impacts of elephants on biodiversity in the Eastern Cape consequences for biodiversity conservation. Biol. Conserv. 163,13–21 (2013). Subtropical Thickets. S. Afr. J. Sci. 102, 395–402 (2006). 40. C. C. Gates, C. H. Freese, P. J. Gogan, M. Kotzman, American Bison: Status Survey and Conser- 14. E. J. Milner-Gulland, E. L. Bennett; The SCB 2002 Annual Meeting Wild Meat Group, Wild vation Guidelines 2010 (IUCN, Gland, Switzerland, 2010). meat: The bigger picture. Trends Ecol. Evol. 18, 351–357 (2003). 41. S. Rule, B. W. Brook, S. G. Haberle, C. S. M. Turney, A. P. Kershaw, C. N. Johnson, The 15. J. S. Brashares, B. Abrahms, K. J. Fiorella, C. D. Golden, C. E. Hojnowski, R. A. Marsh, aftermath of megafaunal extinction: Ecosystem transformation in Pleistocene Australia. D.J.McCauley,T.A.Nuñez,K.Seto,L.Withey,Wildlifedeclineandsocialconflict. Science 335, 1483–1486 (2012). Science 345,376–378 (2014). 42. J. L. Gill, J. W. Williams, S. T. Jackson, K. B. Lininger, G. S. Robinson, Pleistocene megafaunal 16. I. D. Craigie, J. E. M. Baillie, A. Balmford, C. Carbone, B. Collen, R. E. Green, J. M. Hutton, collapse, novel plant communities, and enhanced fire regimes in North America. Science Large mammal population declines in Africa’s protected areas. Biol. Conserv. 143, 2221–2228 326, 1100–1103 (2009). (2010). 43. C. J. Sandom, R. Ejrnaes, M. D. D. Hansen, J.-C. Svenning, High herbivore density asso- 17. P. A. Lindsey, G. Balme, M. Becker, C. Begg, C. Bento, C. Bocchino, A. Dickman, R. W. Diggle, ciated with vegetation diversity in interglacial ecosystems. Proc. Natl. Acad. Sci. U.S.A. 111, H. Eves, P. Henschel, D. Lewis, K. Marnewick, J. Mattheus, J. W. McNutt, R. McRobb, N. Midlane, 4162–4167 (2014). J. Milanzi, R. Morley, M. Murphree, V. Opyene, J. Phadima, G. Purchase, D. Rentsch, C. Roche, 44. H. S. Clements, C. J. Tambling, M. W. Hayward, G. I. H. Kerley, An objective approach to J. Shaw, H. van der Westhuizen, N. Van Vliet, The bushmeat trade in African savannas: determining the weight ranges of prey preferred by and accessible to the five large African Impacts, drivers, and possible solutions. Biol. Conserv. 160,80–96 (2013). carnivores. PLOS One 9, e101054 (2014). 18. W. J. Ripple, B. Van Valkenburgh, Linking top-down forces to the Pleistocene megafaunal 45. A. J. Loveridge, J. E. Hunt, F. Murindagomo, D. W. Macdonald, Influence of drought on extinctions. BioScience 60, 516–526 (2010). predation of elephant (Loxodonta africana) calves by lions (Panthera leo) in an African 19. R. Dirzo, H. S. Young, M. Galetti, G. Ceballos, N. J. B. Isaac, B. Collen, Defaunation in the wooded savannah. J. Zool. 270, 523–530 (2006). Anthropocene. Science 345, 401–406 (2014). 46. C. Carbone, G. M. Mace, S. C. Roberts, D. W. Macdonald, Energetic constraints on the diet 20. T. Levi, G. H. Shepard Jr., J. Ohl-Schacherer, C. A. Peres, D. W. Yu, Modelling the long-term of terrestrial carnivores. Nature 402, 286–288 (1999). sustainability of indigenous hunting in Manu National Park, Peru: Landscape-scale manage- 47. S. R. Loarie, C. J. Tambling, G. P. Asner, Lion hunting behaviour and vegetation structure ment implications for Amazonia. J. Appl. Ecol. 46,804–814 (2009). in an African savanna. Anim. Behav. 85, 899–906 (2013).
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 10 of 12 REVIEW
48. C. J. Tambling, L. Minnie, J. Adendorff, G. I. H. Kerley, Elephants facilitate impact of large 78. J. E. Fa, D. Currie, J. Meeuwig, Bushmeat and food security in the Congo basin: Linkages predators on small ungulate prey species. Basic Appl. Ecol. 14, 694–701 (2013). between wildlife and people’s future. Environ. Conserv. 30,71–78 (2003). 49. L. M. Pereira, R. N. Owen-Smith, M. Moléon, Facultative predation and scavenging by 79. P. A. Lindsey, R. Alexander, M. G. L. Mills, S. Romañach, R. Woodroffe, Wildlife viewing mammalian carnivores: Seasonal, regional and intra guild comparisons. Mammal Rev. preferences of visitors to protected areas in South Africa: Implications for the role of 44,44–55 (2013). ecotourism in conservation. J. Ecotourism 6,19–33 (2007). 50. C. C. Wilmers, W. M. Getz, Gray wolves as climate change buffers in Yellowstone. PLOS 80. R. Naidoo, G. Stuart-Hill, L. C. Weaver, J. Tagg, A. Davis, A. Davidson, Effect of diversity of Biol. 3, e92 (2005). large wildlife species on financial benefits to local communities in northwest Namibia. 51. K. U. Karanth, J. D. Nichols, N. S. Kumar, W. A. Link, J. E. Hines, Tigers and their prey: Environ. Resour. Econ. 48, 321–335 (2011). Predicting carnivore densities from prey abundance. Proc. Natl. Acad. Sci. U.S.A. 101, 81. D. W. S. Challender, D. C. MacMillan, Poaching is more than an enforcement problem. 4854–4858 (2004). Conserv. Lett., in press. 52. P. Henschel, L. Coad, C. Burton, B. Chataigner, A. Dunn, D. MacDonald, Y. Saidu, L. T. B. Hunter, 82. A. N. James, K. J. Gaston, A. Balmford, Balancing the Earth’s accounts. Nature 401,323–324 The lion in West Africa is critically endangered. PLOS One 9, e83500 (2014). (1999). 53. L. P. Rutina, S. R. Moe, J. E. Swenson, Elephant Loxodonta africana driven woodland con- 83. L. Cantú-Salazar, K. J. Gaston, Very large protected areas and their contribution to terrestrial version to shrubland improves dry-season browse availability for impalas Aepyceros melampus. biological conservation. BioScience 60, 808–818 (2010). Wildlife Biol. 11,207–213 (2005). 84.C.Packer,A.Loveridge,S.Canney,T.Caro,S.T.Garnett,M.Pfeifer,K.K.Zander,A.Swanson, 54. M. Landman, G. I. H. Kerley, Elephant both increase and decrease availability of browse D. Macnulty, G. Balme, H. Bauer, C. M. Begg, K. S. Begg, S. Bhalla, C. Bissett, T. Bodasing, resources for black rhinoceros. Biotropica 46,42–49 (2014). H. Brink, A. Burger, A. C. Burton, B. Clegg, S. Dell, A. Delsink, T. Dickerson, S. M. Dloniak, 55. R. Verweij, J. Verrelst, P. E. Loth, I. Heitkönig, A. Brunsting, Grazing lawns contribute to D.Druce,L.Frank,P.Funston,N.Gichohi,R.Groom,C.Hanekom,B.Heath,L.Hunter,H.H.Deiongh, the subsistence of mesoherbivores on dystrophic savannas. Oikos 114, 108–116 C. J. Joubert, S. M. Kasiki, B. Kissui, W. Knocker, B. Leathem, P. A. Lindsey, S. D. Maclennan, (2006). J.W.McNutt,S.M.Miller,S.Naylor,P.Nel,C.Ng’weno,K.Nicholls,J.O.Ogutu,E.Okot-Omoya, 56. M. Landman, D. S. Schoeman, G. I. H. Kerley, Shift in black rhinoceros diet in the presence B. D. Patterson, A. Plumptre, J. Salerno, K. Skinner, R. Slotow, E. A. Sogbohossou, K. J. Stratford, of elephant: Evidence for competition? PLOS One 8, e69771 (2013). C. Winterbach, H. Winterbach, S. Polasky, Conserving large carnivores: Dollars and fence. Ecol. 57. A. Birkett, B. Stevens-Wood, Effect of low rainfall and browsing by large herbivores on an Lett. 16,635–641 (2013). enclosed savannah habitat in Kenya. Afr. J. Ecol. 43, 123–130 (2005). 85. G. S. M. Andrade, J. R. Rhodes, Protected areas and local communities: An inevitable 58. D. H. Janzen, P. S. Martin, Neotropical anachronisms: The fruits the gomphotheres ate. partnership toward successful conservation strategies? Ecol. Soc. 17, 14 (2012). Science 215,19–27 (1982). 86. P. Budhathoki, Linking communities with conservation in developing countries: Buffer 59. A. Campos-Arceiz, S. Blake, Megagardeners of the forest—The role of elephants in seed zone management initiatives in Nepal. Oryx 38, 334–341 (2004). dispersal. Acta Oecol. 37, 542–553 (2011). 87. E. Dinerstein, A. Baccini, M. Anderson, G. Fiske, E. Wikramanayake, D. McLaughlin, G. Powell, 60.Y.Zhou,C.Newman,Z.Xie,D.W.Macdonald,Peduncleselicitlarge-mammalendozoochory D. Olson, A. Joshi, Guiding agricultural expansion to spare tropical forests. Conserv. Lett., in a dry-fruited plant. Ann. Bot. 112,85–93 (2013). in press. 61. S. Blake, S. L. Deem, E. Mossimbo, F. Maisels, P. Walsh, Forest elephants: Tree planters of 88. N. Alexandratos, J. Bruinsma, World Agriculture Towards 2030/2050: The 2012 Revision the Congo. Biotropica 41, 459–468 (2009). (2012); http://environmentportal.in/files/file/World%20agriculture%20towards%202030.pdf. 62. E. Dinerstein, Effects of Rhinoceros unicornis on riverine forest structure in lowland Nepal. 89. C. D. Thomas, A. Cameron, R. E. Green, M. Bakkenes, L. J. Beaumont, Y. C. Collingham, B. F. Erasmus, Ecology 73, 701–704 (1992). M. F. De Siqueira, A. Grainger, L. Hannah, L. Hughes, B. Huntley, A. S. Van Jaarsveld, G. F. Midgley, 63. M. E. Rogers, B. C. Voysey, K. E. Mcdonald, R. J. Parnell, C. E. G. Tutin, Lowland gorillas and L. Miles, M. A. Ortega-Huerta, A. T. Peterson, O. L. Phillips, S. E. Williams, Extinction risk from seed dispersal: The importance of nest sites. Afr. J. Primatol. 45,45–68 (1998). climate change. Nature 427,145–148 (2004). 64. G. O’Farrill, M. Galetti, A. Campos-Arceiz, Frugivory and seed dispersal by tapirs: An 90. S. M. Brook, N. Dudleyb, S. P. Mahood, G. Polet, A. C. Williams, J. W. Duckworth, T. Van Ngoc, insight on their ecological role. Integr. Zool. 8,4–17 (2013). B. Long, Lessons learned from the loss of a flagship: The extinction of the Javan rhinoceros 65. E. O. Effiom, G. Nuñez-Iturri, H. G. Smith, U. Ottosson, O. Olsson, Bushmeat hunting Rhinoceros sondaicus annamiticus from Vietnam. Biol. Conserv. 174,21–29 (2014). changes regeneration of African rainforests. Proc.R.Soc.BBiol.Sci.280, 20130246 91. T. J. Braje, J. M. Erlandson, Human acceleration of animal and plant extinctions: A (2013). Late Pleistocene, Holocene, and Anthropocene continuum. Anthropocene 4,14–23 66. D. Beaune, B. Fruth, L. Bollache, G. Hohmann, F. Bretagnolle, Doom of the elephant- (2013). dependent trees in a Congo tropical forest. Forest Ecol. Manag. 295, 109–117 (2013). 92. M. A. Cozzuol, C. L. Clozato, E. C. Holanda, F.H.G.Rodrigues,S.Nienow,B.deThoisy, 67. J. R. Poulsen, C. J. Clark, T. M. Palmer, Ecological erosion of an Afrotropical forest and R. A. F. Redondo, F. R. Santos, A new species of tapir from the Amazon. J. Mammal. 94, potential consequences for tree recruitment and forest biomass. Biol. Conserv. 163, 1331–1345 (2013). 122–130 (2013). 93. S. T. Turvey, C. T. Trung, V. D. Quyet, H. V. Nhu, D. V. Thoai, V. C. A. Tuan, D. T. Hoa, K. Kacha, 68. C. E. Doughty, A. Wolf, Y. Malhi, The legacy of the Pleistocene megafauna extinctions on T. Sysomphone, S. Wallate, C. T. T. Hai, N. V. Thanh, N. M. Wilkinson, Interview-based nutrient availability in Amazonia. Nat. Geosci. 6, 761–764 (2013). sighting histories can inform regional conservation prioritization for highly threatened 69. J. Pastor, Y. Cohen, N. T. Hobbs, in Large Herbivore Ecology, Ecosystem Dynamics and Con- cryptic species. J. Appl. Ecol. 52,422–433 (2015). servation (Cambridge Univ. Press, Cambridge, 2006), pp. 289–325. 94. IUCN, Saving Saola from Snares (2013); http://iucn.org/news_homepage/news_by_date/? 70. C. Melis, N. Selva, I. Teurlings, C. Skarpe, J. D. C. Linnell, R. Andersen, Soil and vegetation 11769/Saving-Saola-from-snares. nutrient response to bison carcasses in Białowieża Primeval Forest, Poland. Ecol. Res. 22, 95. Convention on Biological Diversity, Aichi Biodiversity Targets (2014); http://cbd.int/sp/targets/. 807–813 (2007). 96. A. Dobson, L. Lynes, How does poaching affect the size of national parks? Trends Ecol. 71. M. S. Waldram, W. J. Bond, W. D. Stock, Ecological engineering by a mega-grazer: White Evol. 23, 177–180 (2008). rhino impacts on a South African savanna. Ecosystems 11, 101–112 (2008). 97. G. Ceballos, P. R. Ehrlich, Mammal population losses and the extinction crisis. Science 296, 72. R. M. Holdo, A. R. E. Sinclair, A. P. Dobson, K. L. Metzger, B. M. Bolker, M. E. Ritchie, R. D. Holt, 904–907 (2002). A disease-mediated trophic cascade in the Serengeti and its implications for ecosystem 98. N. Myers, A. H. Knoll, The biotic crisis and the future of evolution. Proc. Natl. Acad. Sci. U.S.A. C. PLOS Biol. 7, e1000210 (2009). 98,5389–5392 (2001). 73. R. G. Ruggiero, Interspecific feeding associations: Mutualism and semi-parasitism be- 99. Zoological Society of London, EDGE (2014); http://edgeofexistence.org/. tween hippopotami Hippopotamus amphibius and African jacanas Actophilornis africanus. 100. R. M. Pringle, Elephants as agents of habitat creation for small vertebrates at the patch Ibis 138, 346–348 (1996). scale. Ecology 89,26–33 (2008). 74. A. Campos-Arceiz, Shit happens (to be useful)! Use of elephant dung as habitat by amphi- 101. T. S. McCarthy, W. N. Ellery, A. Bloem, Some observations on the geomorphologica impact bians. Biotropica 41, 406–407 (2009). of hippopotamus (Hippopotamus amphibious L.) in the Okavango Delta, Botswana. Afr. J. Ecol. 75. F. Keesing, Impacts of ungulates on the demography and diversity of small mammals in 36,44–56 (1998). central Kenya. Oecologia 116, 381–389 (1998). 102. V. Smil, Harvesting the Biosphere: What We Have Taken from Nature (MIT Press, Cambridge, 76. F. Keesing, Cryptic consumers and the ecology of an African savanna. BioScience 50, 205–215 MA, 2012). (2000). 103. M. Hoffmann, J. L. Belant, J. S. Chanson, N. A. Cox, J. Lamoreux, A. S. Rodrigues, J. Schipper, 77. H. S. Young, R. Dirzo, K. M. Helgen, D. J. McCauley, S. A. Billeter, M. Y. Kosoy, L. M. Osikowicz, S. N. Stuart, The changing fates of the world’smammals.Philos. Trans. R. Soc. B Biol. Sci. 366, D. J. Salkeld, T. P. Young, K. Dittmar, Declines in large wildlife increase landscape-level 2598–2610 (2011). prevalence of rodent-borne disease in Africa. Proc. Natl. Acad. Sci. U.S.A. 111,7036–7041 104. K. E. Jones, J. Bielby, M. Cardillo, S. A. Fritz, J. O’Dell,C.D.L.Orme,K.Safi,W.Sechrest,E.H.Boakes, (2014). C.Carbone,C.Connolly,M.J.Cutts,J.K.Foster,R.Grenyer,M.Habib,C.A.Plaster,PanTHERIA:A
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 11 of 12 REVIEW
species-level database of life history, ecology, and geography of extant and recently extinct Parks (term completed 31 March 2015). T.M.N. is supported by Australian-American Fulbright mammals. Ecology 90, 2648 (2009). Commission. C.J.S. is the Director of Wild Business Ltd., UK. Competing interests: The 105. University of Michigan, The Animal Diversity Web (2014); http://animaldiversity.ummz. authors declare that they have no competing interests. umich.edu. 106. F. A. Smith, S. K. Lyons, S. K. M. Ernest, K. E. Jones, D. M. Kaufman, T. Dayan, P. A. Marquet, Submitted 14 November 2014 J. H. Brown, J. P. Haskell, Body mass of late quaternary mammals: Ecological archives Accepted 3 April 2015 E084-094. Ecology 84, 3403 (2003). Published 1 May 2015 107. FAOSTAT, FAOSTAT Database Agric (2014); http://faostat.fao.org/. 10.1126/sciadv.1400103
Acknowledgments: We thank S. Arbogast for creating the graphics and for her efforts Citation: W. J. Ripple, T. M. Newsome, C. Wolf, R. Dirzo, K. T. Everatt, M. Galetti, M. W. Hayward, drawing the animal outlines in the figures. Funding: K.T.E. received funding from Nelson Mandela G. I. H. Kerley, T. Levi, P. A. Lindsey, D. W. Macdonald, Y. Malhi, L. E. Painter, C. J. Sandom, Metropolitan University, Panthera Kaplan Award. G.I.H.K. is supported (salary) by Nelson Mandela J. Terborgh, B. Van Valkenburgh, Collapse of the world’s largest herbivores. Sci. Adv. 1, Metropolitan University and is a board member (non-executive) of South African National e1400103 (2015).
Ripple et al. Sci. Adv. 2015;1:e1400103 1 May 2015 12 of 12 Tom Rooney, Ramana Callan, Krystle Bouchard, and Nate Nibbelink Summary of two research projects
Ramana Callan, PhD Krystle Bouchard, MS University of Georgia Wright State University Wisconsin’s Understory Plant Communities
• Local losses in plant species diversity
• Regional recruitment failure of conifers
• N. WI White-tailed deer populations - pre-settlement: < 10/mi2 - current: 10-40/mi2 Wisconsin’s Wolves
• Predicted to contribute to the conservation of regional biodiversity
• Through direct impacts on white- tailed deer, wolves are predicted to trigger additional indirect impacts on plant communities Wolves and Trophic Cascades Wolves and Trophic Cascades
Predators can indirectly influence plants by: • Predation: Density-mediated indirect interactions (Death Effects) • Predation Risk: Trait-mediated indirect interactions →Behaviorally mediated trophic cascades (Fear Effects) Wolf-moose-balsam fir system on Isle Royale
(Photo credit: Michigan McLaren and Peterson 1994 Technological University) Trophic interactions in Wisconsin forests Wolves and white-tailed deer
• 400,000 deer; 690 wolves regional Death Effects unlikely • Distribution of deer in MN found to be at margins of wolf territories → buffer zones between packs act as refugia Death and Fear Effects possible • Wolves are predicted to alter foraging behavior by white-tailed deer (i.e. deer increase vigilance and movement) Fear Effects possible
Callan: Northern white cedar wetlands
• High Plant Species Diversity • Historically used by deer as winter “yards” Objective
Detect and characterize differences in vegetation between areas occupied by wolves and areas unoccupied by wolves Experimental Approach
Overlay Wisconsin DNR wolf territory data - characterize “high wolf impact areas” (8-10 years of wolf occupancy) and “low wolf impact areas” (0-3 years of wolf occupancy)
Research Questions
(1) Is plant species richness higher in white cedar wetlands occupied by wolves? - species richness by vegetation growth form: Tree, Shrub, Forb, Fern, Grass (2) At what scale are these differences detectable? - 0.01m², 0.1m², 1m², 10m², 100m², 1,000m² Hypothesis: Spatial Scale
Species Area Curve Indicating Higher Local Effects
High Potential 80 Wolf Impact 70 60 Low Potential 50 Wolf Impact 40 30 20 Species Richness 10 0 Site Location Ecosystem Landscape Region Spatial Scale Hypotheses: Vegetation Growth Forms
High Impact Wolf Areas should display: 1) ≈ Tree species richness and % cover 2) ↑ Shrub and Forb species richness Wolf Recovery 3) ↓ Grass and Fern % browsing intensity cover n=38 0.01m² Carolina Vegetation 0.1m² Survey (CVS) Protocol 1m²
10m² Plot Diagram Results: Biodiversity of white cedar wetlands
Trees 23 Shrubs 31 Forbs and Vines 100 Ferns and Fern allies 17 Sedges 16 Grasses and Rushes 8 Non-natives 4 Hypothesis 1 Supported: ≈ Tree species richness and % cover
Tree 100
80
60
40
Percent Cover 20
0
Wolf
Non_wolf Hypothesis 2 Supported: ↑ Forb & Shrub richness in wolf areas Hypothesis 2 Supported: ↑ Forb & Shrub richness in wolf areas Results: Percent Cover
Hypothesis 3 Partially Supported: ↓ Grass and Fern % cover in high wolf impact areas
Grass 2.0 Fern 30 * 1.5 20 1.0
10
Percent Cover 0.5
0 0.0
Wolf
Low wolf High wolf Wolf LowNon_wolf wolf High wolf Non_wolf Low wolf impact area High wolf impact area Response of browse-sensitive species in low and high wolf impact areas Results: Select sensitive species
Wild Sarsaparilla
2 40 24.7± 8.4
Wild sarsaparilla 30 (Aralia nudicaulis) 20
4.7± 1.6 10
Average abundance/ 100m 0
Low wolf HighWolf wolf Non_wolf Nodding Trillium
Nodding trillium 2 2.5 1.40 ± 0.62 (Trillium cernuum) 2.0
1.5
1.0 0.43 ± 0.22
0.5
Average abundance/ 100m 0.0
Low wolf HighWolf wolf Non_wolf Bouchard: Upland forest wildflowers
• Focus on 3 deer browse indicator species Objective
Detect and characterize differences in vegetation across a wolf recolonization gradient Research Questions
(1) Does indicator plant size increase with time since wolf recolonization?
(2) How long does it take before wolf effects become detectable? Experimental Approach
(1) Overlay Wisconsin DNR wolf territory data - wolves present 12-13 years - wolves present 4-6 years - wolves absent (2) Sites on national and state forest land; matched stand types (mature forest)
Hypothesis: Plant Size
(1) Mean indicator plant size increase with time since wolf recolonization, but does not resemble “deer-free” exclosures Results: Mixed Effects After 4-6 Years, Consistent Effects After 12-13 Years Results: Mixed Effects After 4-6 Years, Consistent Effects After 12-13 Years Summary of Results
• Species richness of forbs and shrubs was greater in high wolf impact areas and evident at specific scales: » 1m²-10m² for forbs » 10m²-400m² for shrubs • % cover of ferns was lower in high wolf impact areas • Browse indicator species reveal reduced browsing pressure in high wolf impact areas Summary of Results
• In forests and forested wetlands, trophic cascades: – Exist – Are subtle – Require about a decade before they are apparent – Do not resemble “deer free” conditions – Might become more pronounced with time Acknowledgements
Collaborators Adrian Wydeven, Jane Wiedenhoeft (Wisconsin DNR) Corey Raimond and Clare Frederick (Field Assistants) Warren Keith Moser (Forest Service) ...
DEC 19 1977
.. . ,. various forage improvement schemes, and a host of other manage- ment practices. The improved management practices used in the ··.~+.·. private sector are the same kinds of improvements needed to develop and manage the public rangelands.
Consideration of Predator Losses and Other Adverse Impacts on Animal Production Permittees believe predator losses are greater on public land than on private land. Reasons are centered around Government failure to use an effective coyote control program and the nature of the range landscape which makes the land a natural habitat for predators. Predator losses are identified as result ing from coyotes, bears, bobcats, lions, and feral dogs. However, most of the comments concern the predator losses resulting from an increased coyote population. Since permittees believe predator losses are greater on public land, they feel there should be an increase In cost allowance for predatio~ i.e., a fee reduction. Predator losses have been given consideration in one item of the Federal land user costs - death loss of animals. A difference was recognized in the larger number of death loaeee,on public land and this larger loss was reflected in a lower FMV for public land. The cost difference between public and private land was the second largest, $0.23 per AUM. The only justification for a higher death loss cost allowance would be for any recent change in federal land predator losses that is significantly different from predator losses on private lands. Available data do not support arguments that predator losses have increased faster on Federal lands than on private lands, but th•t death losses have increased on both. An estimate could be made of added rancher costs due to predators, with additional reduction being given in the grazing fee. This approach would require determination of predator costs on an individual rancher or limited area basis.
Credit for Wildlife Use of'Private Lands The proposal for a variable fee system for grazing on Federal lands is based in part on t he presence of wildlife as well as recreational uses of associated Federal, State, and Private lands. The benefits of wildlife and recreation accrue in these areas to the individual recreationists as well as the general public. However, part of the cost for support of this wildlife may be borne disproportionately by individual landowners relative to their benefits from wildlife and recreation.
3-21 Uniform Bail and Penalty Schedules
JULY 2011 EDITION
(Cal. Rules of Court, rule 4.102)
TRAFFIC BOATING FORESTRY FISH AND GAME PUBLIC UTILITIES PARKS AND RECREATION BUSINESS LICENSING
/~ JUDICIAL COUNCIL ~y;V OF CALIFORNIA
-- ~~-----~---~-~ FISH AND GAME BAIL AND PENALTY SCHEDULE (*See Preface, Section III(B)) (**See Preface, Section IV)
OJ ~ EMS .2 ..., Base State County DNA Court ..::: PA* Total ~ "' > .. 0 ~ OJ ::l"' 0 Section ~ ....:: Offense Fine PA PA/10 PA* PA*/10 BURRO 4600 (a) M Unlawful Killing or Capture of Undomesticated 300 300 210.00 120 150.00 60 60 1,200.00 + Burro WILD PIG 4657 M Wild Pig License Tag Violation 150 150 105.00 60 75.00 30 30 600.00 + 4669 M Failure to Secure Liability Insurance Protection by 450 450 315.00 180 225.00 90 90 1,800.00 + For-Hire Vessel Operator FULLY PROTECTED MAMMALS 4700 (a)(l) 7 M Taking or Possession of Fully Protected Mammal 5,000 5,000 3,500.00 2,000 2,500.00 1,000 1,000 20,000.00 + BEAR 4753 M Bear Tag License Violation 150 150 105.00 60 75.00 30 30 600.00 + 4758 M Sale or Purchase of Bear Meat or Parts 3,250 3,250 2,275.00 1,300 1,625.00 650 650 13,000.00 + 4758 M Possession of Three or More Bear Gall Bladders 10,000 10,000 7,000.00 4,000 5,000.00 2,000 2,000 40,000.00 + FULLY PROTECTED REPTILES AND AMPHIBIANS 5000 M Selling, Purchasing, Harming, Possessing, 300 300 210.00 120 150.00 60 60 1,200.00 + Transporting, or Shooting Desert Tortoises 5050 (a) M Taking or Possession of Fully Protected Reptile or 500 500 350.00 200 250.00 100 100 2,000.00 + Amphibian FISH-GENERAL PROVISIONS 5508 10 M Fish oflndeterminate Size 250 250 175.00 100 125.00 50 50 1,000.00 + (For Each Additional Fish) 20 20 14.00 8 10.00 4 4 80.00 + + $40 court security fee (reduced to $30 on 7/1/13 unless sunset repealed) and conviction assessment ($35/infraction; $3D/misdemeanor) per offense. 121 T-Shirt Weather in the Arctic - NYTimes.co Page 1 of 5 from $119.00 Opinion SUBSCRIBE | LOG IN OP-ED CONTRIBUTORS T-Shirt Weather in the Arctic By MARK URBAN and LINDA DEEGAN FEBRUARY 5, 2016 WE crested the northern rim of Alaska’s Brooks Range, and from the windows of our truck looked out across the undulating foothills toward the Arctic Ocean. Instead of seeing snow as we had in years past, we were greeted by a landscape already green with spring. We flew by helicopter to our remote camp and shed our heavy parkas. The fish we had come to study had already disappeared downstream to spawn. We now realize that what we saw last May was historic — the hottest May for Alaska’s North Slope during what scientists recently concluded was the hottest year on record for the earth. We also saw the future. Last year, the earth’s temperature passed the mark of 1 degree Celsius above preindustrial levels. Civilization took 165 years to reach that mark, and now the increase could reach 2 degrees Celsius in just 30 more years, a point at which the risks from sea-level rise, drought and other effects could increase significantly. Arctic Grayling NATIONAL GEOGRAPHIC CREATIVE http://mobile.nytimes.com/2016/02/06/opinion/t-shirt-weather-in-the-arctic.html?_r=0 2/10/2016 T-Shirt Weather in the Arctic - NYTimes.co Page 2 of 5 Despite promises made in Paris to cut greenhouse gas emissions, we will still need to make it through the hottest years of a looming global heat age. Along with the many challenges we face, we must figure out how to protect ecosystems and the benefits they provide. Each spring for the last 30 years, our team of biologists has traveled to remote field camps in Arctic Alaska. The Arctic is warming faster than anywhere else in the world as seawater replaces sea ice, painting the Arctic Ocean blue and fueling a dangerous feedback loop. The white sea ice reflects the sun’s energy back into space through what is known as the albedo effect. But as the ice melts, the dark Arctic seawater is now absorbing that heat, turning up the earth’s temperature. With the early spring, snow melted roughly two weeks earlier than in the past and plants turned green soon after. Lakes thawed about 10 days earlier, and Arctic grayling, a fish, bred weeks earlier. An early spring has long-term consequences. When grayling breed three weeks earlier, for instance, their offspring get a head start on feeding and grow nine times larger. This might seem like a good thing, until you consider that the same warmer temperatures dry the rivers that enable these grayling to swim to lakes where they spend the winter. As these fish wait in shallow pools for the rivers to flow, bears and birds enjoy a captive feast. If rivers do not flow before winter, the fish freeze. The drying of these rivers could threaten some grayling populations. Last May’s warmth deceived white-crowned sparrows into breeding earlier than usual. When a snowstorm roared in, the sparrows abandoned their ill-timed nests, leaving their eggs behind to perish. Thunderstorms also raged over our camp. These storms used to be rare in the Arctic, but they strike often now. Lightning has set fire to the tundra, releasing into the atmosphere huge stores of ancient carbon from the permafrost. Sinkholes are also opening up in the thawing tundra. Walk up to one, and you will hear the trickle and clatter as heat dissolves permafrost into cascades of ice age mud and stones. We are only just beginning to understand these changes. Ecosystems involve a complex web of connections among species and the physical environment. Climate change alters these connections in ways that can surprise and baffle us. For example, scientists thought they understood Arctic streams until we added nutrients to one to mimic what happens when the tundra thaws. A rare moss materialized and blanketed the streambed. A new set of insects appeared, but they http://mobile.nytimes.com/2016/02/06/opinion/t-shirt-weather-in-the-arctic.html?_r=0 2/10/2016 T-Shirt Weather in the Arctic - NYTimes.co Page 3 of 5 sheltered in the moss instead of drifting into the waiting mouths of hungry grayling. So in a roundabout way, a more productive stream made for skinnier fish. The surprises pose serious risks because we can’t prepare for what we don’t know. We can no longer be satisfied to watch and document these changes. We must predict and prevent them. Sustaining life through the coming heat age will require tough decisions as we triage the rising number of climate casualties. We cannot hope to save all species when we haven’t even figured out how many species there are. We might focus initially on protecting those with the greatest importance to other species and ecosystems, the so-called biotic multipliers of climate change. For instance, top predators are often sensitive to climate change and magnify climate effects by yanking hard on the threads that connect them to other species in the food chain. Our current approaches to identifying which species and ecosystems are most at risk are primitive. Most predictions rely on the correlation between a map of an animal’s range and a few climate factors. As biologists, we need to develop forecasts that rely on causes, not correlations, as our colleagues studying the atmosphere did years ago. This will require an enhanced effort to comprehend how species survive, reproduce, evolve and move across landscapes, and how changes in the climate alter each of these factors. We also need experiments that replicate a warming environment. Scientists know how to heat small plots of tundra with open-topped plastic enclosures and forests with heated cables. But the small size of these efforts limits our ability to understand consequences for larger animals and ecosystems. We need to engineer ways to warm bigger ecosystems experimentally by heating up entire lakes, streams, fields and even forests. We plan to return to the Arctic again in May. This year is predicted to be even hotter than the last. We’ll be ready this time. We understand now that we have already entered the heat age. Mark Urban is an associate professor in ecology and evolutionary biology at the University of Connecticut in Storrs. Linda Deegan is a senior scientist at the Marine Biological Laboratory Ecosystems Center in Woods Hole, Mass. Follow The New York Times Opinion section on Facebook and Twitter, and sign up for the Opinion Today newsletter. http://mobile.nytimes.com/2016/02/06/opinion/t-shirt-weather-in-the-arctic.html?_r=0 2/10/2016 Making Room for Wolf Recovery: The Case for Maintaining Endangered Species Act Protections for America’s Wolves ildlife for W 25 g Y in e v a a r S s Amaroq Weiss, Noah Greenwald, Curt Bradley November 2014 I. Executive Summary southern Rocky Mountains, Grand Canyon, Cascade Mountains in Washington, Oregon and California, In 2011 the U.S. Fish and Wildlife Service removed the Sierra Nevada and the Adirondacks are all places Endangered Species Act protections for wolves in the that could support wolf populations. According to northern Rocky Mountains and western Great Lakes, the studies, these areas are capable of supporting a arguing that wolves were recovered in those regions minimum of 5,000 wolves, which would nearly double and the states could be trusted to manage them. But the existing wolf population. all of the states with substantial wolf populations have enacted aggressive hunting and trapping seasons that Recovering wolves to these additional areas is are intended to drastically reduce wolf populations. To necessary to ensure the long-term survival of gray date these hunts have resulted in the killing of more wolves in the lower 48 states and enrich the diversity than 2,800 wolves. The deaths of so many wolves of U.S. ecosystems that have lacked the gray wolf have contributed to declines in wolf populations of as a top predator for decades. At last count the three 9 percent in the northern Rockies and 25 percent in existing wolf populations combined include only Minnesota. Given increased efforts to kill wolves roughly 5,400 wolves, which is below what scientists in many states, these declines can be expected to have identified as the minimum viable population continue and likely increase. size necessary to avoid extinction. Considering that populations are now declining and isolated at several Despite the nightmare that state management of wolves scales, doubling the population by facilitating wolf has been, the Fish and Wildlife Service has proposed to recovery in additional areas is needed to secure the remove protections for gray wolves in the remainder of future of gray wolves in the U.S. the lower 48 states, excluding a small portion of Arizona and New Mexico, where the Mexican gray wolf struggles Studies following reintroduction of wolves to to survive. The agency argues that growth of populations Yellowstone National Park have documented that in the northern Rockies and Great Lakes is sufficient wolves as top predators play pivotal roles in shaping to consider the species recovered and to remove the structure and function of ecosystems, benefitting Endangered Species Act protections. a wide range of species, including beavers, songbirds, grizzly bears, foxes, bison, pronghorn and more. In this report, we make the case that the job of recovering wolves is far from complete by: Gray wolves are also a substantial draw for people from around the world. Millions of people have traveled to • Identifying and mapping suitable habitat not Yellowstone from around the world to see the gray currently occupied by wolves; wolves reintroduced in 1995 and 1996, and polls • Documenting dispersals of wolves to this consistently show that a broad majority of the American habitat; public supports the recovery of gray wolves, including • Detailing the limitations of current to new areas where they don’t currently occur. management plans; • Highlighting the important roles wolves play in ecosystems. Gray wolves currently occupy less than 10 percent of their historic range and a fraction of currently suitable habitat. To identify and map unoccupied, suitable wolf habitat in the United States, we used 27 studies that model wolf habitat in different regions to create a single map. Based on this analysis, there is up to 530,000 square miles of suitable wolf habitat in the United States, only roughly 171,000 square miles of which is occupied, demonstrating that wolves currently occupy only about 30 percent of existing suitable habitat. The Wenaha pack male courtesy ODFW 1 Cover photo by Chris Smith / Flickr II. Introduction The agency is now proposing to remove protections for wolves across the lower 48 excluding a portion Gray wolves once occupied the majority of North of the range of the Mexican gray wolf in Arizona and America, excluding perhaps only the driest deserts and New Mexico (FWS 2013). This proposal disregards the southeastern United States, where the red wolf occurs that there are only roughly 5,400 wolves in portions (FWS 2013). Scientists estimate that pre-European of the Midwest (~3,700 wolves), northern Rockies settlement there may have been as many as 2 million (~1,670) and Southwest (~80) (FWS 2013), and the wolves in North America (Leonard et al., 2005). During states are actively working to reduce populations. the late 19th century and early 20th century, state and Moreover, wolves occupy just a fraction of their local bounties reduced wolf numbers. From 1915 historic range, less than 10 percent, and only a small through mid-century, the U.S. government exterminated portion of existing suitable habitat. Indeed, multiple wolves from the United States and Mexico (Seton, 1929; researchers have modeled extensive suitable habitat Young and Goldman, 1944). By 1967, when wolves for wolves in the Northeast, Pacific Northwest, were protected under a precursor to the Endangered southern Rocky Mountains, California and elsewhere. Species Act, they had been reduced to fewer than 1,000 To describe the full extent of suitable habitat available wolves in northeastern Minnesota (FWS 2009). for further recovery of wolves, we reviewed literature estimating existing wolf habitat, created composite With protection, wolves began to see some recovery, maps of all known wolf habitat in the lower 48, but only in portions of their former range where the quantified unoccupied habitat, and estimated the U.S. Fish and Wildlife Service (FWS) developed minimum number of wolves that could occur in this recovery programs. Wolves were originally protected habitat. We also quantified and mapped wolf dispersal as four subspecies -- the northern Rockies wolf, events over the past 30 years. In the following eastern wolf, Mexican wolf and Texas wolf (FWS discussion, we present the results of these analyses, 1978). Recognizing that these subspecific designations further discuss the history of efforts to remove were potentially invalid, FWS consolidated protection protections for wolves, including discussion of the for gray wolves to the species level in 1978, including current proposal, and provide a rationale for not the entire lower 48 states (Ibid.) The agency, however, walking away from wolf recovery now. never developed a national strategy to recover wolves in the lower 48 in line with expanded protection. III. Studies Estimating Wolf Habitat in the Instead it completed recovery plans that had already United States been started in 1978 for three of the four purported subspecies, excluding the Texas wolf. We reviewed 27 studies that modeled potential wolf habitat in the lower 48 states and used the composite With recovery programs in place, including results to estimate and map the full range of potential reintroduction of wolves in portions of the northern unoccupied wolf habitat and the number of wolves that Rocky Mountains, wolves began to grow in number could be supported in the lower 48 (Appendix A). The and expand their range in the northern Rockies and studies modeled wolf habitat across the western United western Great Lakes states. Mexican wolves were States, the upper Midwest and the Northeast (Appendix also reintroduced to a portion of the Southwest, but B). This likely encapsulates a majority of existing wolf their numbers have grown slowly. In 2003 FWS began habitat in the United States excluding the range of the moving to delist wolves in the northern Rockies and red wolf in the Southeast. But there may be additional western Great Lakes, and after multiple rounds of habitat in North and South Dakota and other areas that litigation in which the agency was repeatedly found should be the subject of additional modeling. not to have followed best science, were successful in removing protections in both regions in 2011 (FWS Predictive modeling parameters used in the studies 2011ab). Since delisting, all states in the northern consisted of road density (26 studies), human Rockies and western Great Lakes have instituted population density (20 studies), prey density (20 aggressive hunting and trapping seasons intended to studies) and land cover/use (16 studies). Some drastically reduce wolf populations. studies used additional parameters including 2 land ownership (11 studies), livestock density (7 and the Sierra Nevada in California. In the northeastern studies), slope or elevation (5 studies), climate or United States, thousands of square miles of terrain snowfall (4 studies), surface water availability (4 spanning upstate New York and portions of Vermont, studies), and prey accessibility or availability (3 New Hampshire and Maine were identified as capable of studies). Two studies used soil depth or hydrology supporting a wolf population. And some studies indicate (Appendix C). the lower peninsula of Michigan could support wolves. Figure 1. Suitable gray wolf habitat in the contiguous United States as identified in 14 modeling studies. Past modeling of wolf habitat has accurately predicted wolf occupancy in both the northern Rockies and According to our mapping, there are approximately Midwest, suggesting modeling can accurately convey 530,000 square miles of suitable wolf habitat in the potential wolf habitat. We used 14 of the 27 studies to lower 48, of which roughly 171,000 square miles are create a composite map of wolf habitat for the lower currently occupied, meaning wolves have recovered 48 states (Figure 1). We did not use all of the studies to only roughly 30 percent of known suitable habitat. because in some cases they represented different Although not all studies estimated the number of wolves modeling iterations for the same areas by the same that could be supported, those that did suggest that at authors, and in others there was insufficient spatial least another 5,000 wolves could populate the Northeast, information to allow mapping. southern Rockies, West Coast and Southwest, nearly doubling the existing population and creating a network Reviewed studies identified extensive wolf habitat in of interconnected populations bolstering genetic security. regions where wolves have not yet recovered. In the western United States, this includes the central and IV. Wolves Are Dispersing Into Areas of southern Rocky Mountains in both Colorado and Utah, Suitable Habitat and Need Endangered the Grand Canyon and surrounding areas in northern Species Act Protections to Survive Arizona, the Olympic Peninsula in Washington, the Cascade Mountains in Washington, Oregon and California Not only is there extensive suitable habitat in other 3 regions of the country, but wolves are dispersing into this habitat. Wolves can travel substantial distances traversing diverse landscapes when leaving their birth- packs to seek mates and territory of their own (Mech and Boitani, 2003). The most-recent and well-known example is that of wolf OR- 7, who traveled more than 4,000 miles after dispersing from his birth pack in northeastern Oregon to travel to California and back into Oregon repeatedly during 2011-2014. He recently found a mate, with whom he has denned and produced pups in southwestern Oregon just north of the California border. In order to quantify and visually display these Figure 2. Dispersals by wolves to locations outside dispersal events, we tabulated all known wolf dispersals of core federal recovery areas, 1981-2014. between 1981 and 2014 in which wolves dispersed to areas and states outside of existing core recovery areas that support source populations and in areas to areas (Appendix D). The dispersals we tabulated were which wolves are dispersing. reported in newspaper stories, agency reports and other sources, and for each dispersal event we attempted to V. Recovery of Wolves to Additional Areas is obtain a point of origin and endpoint. We identified Required by the Endangered Species Act 56 dispersal events in total, with an average dispersal distance of 264 miles. This data shows that wolves have Unlike previous endangered species statutes, the and will continue to move into suitable habitat on the Endangered Species Act does not simply require West Coast, southern Rocky Mountains and Northeast, recovery of species to the point that they are not at where they need protection if they are going to survive risk of global extinction. Indeed, the primary purpose and establish populations (Figure 2). Indeed, with of the Act is to conserve the ecosystems upon which protections under the Endangered Species Act, wolves endangered species depend (16 U.S.C. § 1531(b)). were able to move into Oregon and Washington from Significantly, the Act defines an endangered species as both the northern Rockies and British Columbia and any species in danger of extinction in all or a significant form fledgling populations. portion of its range (16 U.S.C § 1532(6)). This means that a species need not be at risk of extinction Our data also shows dispersal events steadily everywhere to qualify for protection, but rather only increased from 2000 to 2011, when populations were in a significant portion of range. Accordingly, it also steadily growing with endangered species protections means that species cannot be considered recovered in place, and appear to have since declined now that until no longer endangered in any significant portion of all states with substantial wolf populations have range. As demonstrated by the 27 studies we reviewed, enacted aggressive hunting and trapping seasons, wolves remain absent or at very low numbers over leading to population declines (Figure 3). This further significant portions of their historic range where there highlights the need for continued protection both in 4 Figure 3. Annual gray wolf dispersal events to areas and states outside of federal core recovery areas. is extensive remaining habitat, including the Northeast, numbered 1,691 in the northern Rockies, 3,669 in southern Rocky Mountains, West Coast and elsewhere. the western Great Lakes, and 83 in the Southwest.i For this reason alone, wolves remain an endangered According to the above studies, which collectively species that continues to need the protections of the examined hundreds of species, all existing wolf Endangered Species Act. populations are below minimum population sizes considered necessary to ensure long-term survival. VI. Existing Wolf Populations Are Not Viable in the Absence of Additional Of further concern, wolf populations in both the Population Expansion northern Rockies and Great Lakes are declining in response to aggressive hunting and trapping The existing wolf populations in the northern Rocky seasons enacted by individual states. In the northern Mountains, western Great Lakes and Southwest are Rockies, the last population count showed a 9 percent below minimum viable population sizes sufficient decline since federal delisting and in Minnesota, to ensure their survival (Shaffer, 1981; Reed et. al., the population declined by an estimated 25 percent 2003, Traill et al. 2007). In an analysis of 102 species between 2008 and 2012.ii If these population declines including the gray wolf, Reed et al., (2003) estimated continue, risk to wolf populations will only increase. a mean and median minimum viable population of 7,316 and 5,816 individuals respectively, concluding Existing wolf populations are also below levels that long-term persistence of wild populations considered necessary to avoid genetic inbreeding. A of animals, such as wolves, requires 7,000 adult number of studies have concluded that an “effective” individuals. Likewise, Traill et al., (2007) combined population size of 500 individuals is necessary to avoid results from studies on 212 species, including the i gray wolf, finding that the median minimum viable For northern Rockies see: http://www.fws.gov/mountain-prairie/ species/mammals/wolf/annualrpt12/; for western Great Lakes see population was 4,169 individuals. http://www.fws.gov/midwest/wolf/aboutwolves/WolfPopUS.htm. As of the end of 2013, the existing wolf populations ii See: Ibid. and www.mndnr.gov/wolves 5 OR-11 courtesy ODFW the effects of genetic inbreeding (Soule and Wilcox, VII. Recovering Wolves to Additional Areas 1980; Frankel and Soule, 1981; Soule, 1986; Franklin Is Necessary for Healthy, Functioning and Frankham, 1998). Effective population size is Ecosystems defined as the number of breeding individuals, rather than total individuals, translating into a total population The loss of large carnivores is a global problem with of 2,500-5,000 individuals to maintain a total of 500 broad ecological consequences. Because of their breeding individuals (Frankham, 1995). Gray wolves position at the top of food chains, large carnivores play in North America have already lost substantial genetic an inordinate role in shaping the structure and function diversity because of the severe reduction in their overall of diverse ecosystems (Estes et al. 2011, Ripple et historical numbers and range and further losses could al., 2014). According to Ripple et al. (2014), nearly lower survival and reproduction further endangering two-thirds (61 percent) of large carnivore species wolves (Leonard et al., 2005). are considered threatened by the IUCN and most (77 percent) are declining. The extirpation of the gray wolf Loss of genetic diversity due to small population across most of the American landscape is no exception size and historic declines is compounded by the to this pattern and as elsewhere around the world, loss isolation of existing wolf populations (Soule, 1980). of a top predator like the gray wolf has resulted in a The Mexican gray wolf, for example, is isolated from number of rippling ecological consequences that have all other wolf populations and the population in the negatively impacted a broad range of species. This can Greater Yellowstone ecosystem is largely isolated be inferred largely by studies showing positive trends from other wolves in the northern Rockies. This lack in a broad range of species following reintroduction of of connectivity further increases the potential for wolves in the northern Rockies. loss of genetic variation over time. Restoring wolves to additional suitable habitat would create more Studies following reintroduction of wolves to opportunity for connectivity between populations. Yellowstone National Park documented that wolves It would also increase the likelihood that wolves had a profound and transformative impact on the dispersing from currently existing wolf populations landscape that benefitted a wide variety of species. In would be able to find other wolves with whom to particular, the reintroduction of wolves resulted in a mate, and thus contribute genetically to the health of dramatic decrease in elk numbers and also potentially adjacent populations. forced them to move more (Barber-Meyer et al., 2008, 6 Diamond pack female courtesy WDFW Ripple and Beschta, 2012). Reduced elk browse in survival of pronghorn antelope fawns due to reduced turn has led to recovery of woody species, such as predation by coyotes (Berger et al., 2008). Carcasses cottonwood, aspen, willow and serviceberry (Ripple of elk killed by wolves provide food for a host of other and Beschta, 2012). This has fostered many beneficial scavenger species, including but not limited to grizzly ecosystem changes, from providing crucial nesting and bears, black bears, coyotes, eagles, ravens, magpies roosting sites for songbirds, to enhancing root strength and hundreds of species of beetles (Smith et al., 2003). and thereby protecting streams from soil erosion, Wolf-kills may also provide a buffering effect against to providing food and building sources for beavers climate change for carrion-feeders that depend on whose dams then create cool, deep ponds needed carcasses for food. As warming temperatures result in by juvenile fish, and finally to facilitating growth of decreased winter severity, and thus a decreased die-off berry-producing shrubs that provide food for grizzly of vulnerable animals that would otherwise succumb bears and other animals (Ripple and Beschta, 2004; to harsh weather, wolf-kills will provide the carcasses Hebblewhite et al., 2005; Weiss et al., 2007; Eisenberg scavengers need to survive (Wilmers and Getz, 2005). et al., 2013, Hollenbeck and Ripple, 2008, Ripple et al., 2013). The ecosystems of the southern Rocky Mountains, Colorado Plateau, Grand Canyon, Cascade Mountains, Wolves prey on wild ungulates which are the most Adirondack Mountains, Sierra Nevada and elsewhere vulnerable due to factors such as age, injury or ill- would all benefit from the return and recovery of health, allowing greater numbers of healthier, more the gray wolf. It is not enough to restore the wolf robust, and more alert animals to survive and pass to small fragments of its historic range. Instead, on their genes (Stahler et al, 2006). Wolves may also large carnivores like wolves should be restored to prevent the spread of diseases among prey species population levels allowing them to once again be by culling sick animals before they infect others “ecologically effective” – that is, a population that (Wild et al., 2005). Wolves view coyotes as territorial has enough individuals and a wide enough geographic competitors and in some parts of Yellowstone wolves distribution so that not just the species’ existence has have greatly decreased the coyote density. This has been reestablished but, also, its essential role in nature led to increases in numbers of foxes and increased (Soule et al., 2003; Carroll et al., 2006.) 7 be pregnant.) In Wyoming, wolves were designated VIII. Maintaining Federal Protections for a predatory animal that can be killed at anytime by Wolves Is Essential Because States Cannot nearly any means, including killing pups in dens, in Be Trusted to Conserve Existing Wolf nearly 85 percent of the state and designated the rest Populations or Protect Wolves Dispersing to of the state outside of Yellowstone and Grand Teton Other Areas National Parks, a trophy game area, where hunting of wolves was permitted. This management was found Following removal of Endangered Species Act to be inadequate by a federal court in September, protections in the northern Rockies and western 2014 and Endangered Species Act protections were Great Lakes, all of the states with substantial reestablished, stopping the 2014-2015 hunt and killing wolf populations enacted aggressive hunting and of wolves in the predatory zone. trapping seasons designed to drastically lower populations, and indeed population declines are In Minnesota, the state had promised in its state occurring. In the three years since protections were wolf plan that there would be no hunting or trapping removed, nearly 3,000 wolves have been killed of wolves for five years post delisting, but instead through state-sanctioned “harvest” seasons. The instituted wolf hunting and trapping immediately killing of so many wolves in such a short time following delisting. To date, at least 650 wolves directly reflects the negative prejudices towards have been killed and the population declined by wolves held by powerful minorities in all of these 25 percent between 2008-2012. Starting in 2012, states. These prejudices were the primary cause Wisconsin authorized wolf-hunting and trapping of the extirpation of the wolf across significant that has to date resulted in killing of 374 wolves portions of its range and highlight why wolves with a goal of reducing the population by more continue to need federal protections and a national than half to 350 wolves from over 800. Wolves recovery plan. are allowed to be hunted and trapped with the use of hounds, night-hunting by artificial lights, and Worse still, anti-wolf policies appear to be getting baiting, despite overwhelming public opposition more severe in most states where protections have to any of these practices.iv Michigan’s governor been removed. In Idaho, for example, wolf hunting in 2013 signed a bill allowing its state department is allowed year round, including during breeding of natural resources to institute hunting of wolves season and has resulted in the death of at least 1,000 despite citizens having collected over a quarter wolves and reduced the state’s wolf population by of a million signatures to place a no-wolf-hunting around 23 percent from its 2008 peak. This not being measure on the election-season ballot; a second enough, the Idaho Department of Fish and Game in signature-collecting effort has resulted in a second January 2014 hired a bounty hunter to pack into the ballot measure to overturn the newly-signed wolf- Frank Church-River-of-No-Return Wilderness where hunting law, but a pro-hunting ballot measure was he killed nine wolves; has sent U.S. Department of just passed by the legislature. Because the no-wolf Agriculture/Wildlife Services’ airborne sharpshooters hunting ballot measure must be decided by the into the Clearwater National Forest where 48 wolves voters in the November election, Michigan wolves have been killed in six operations; and Gov. Butch have received a temporary reprieve and there will be Otter this spring signed into law a bill that establishes no wolf-hunting season in Michigan this year. But if a wolf-control board and provides over $600,000 the no-hunting measure does not pass, the legislative annually to kill most of Idaho’s remaining wolves.iii bill will go into effect in March meaning Michigan’s In Montana, the state wildlife commission nearly wolves will once again be facing legal harvest and doubled the number of wolves that can be killed by certain death. Finally, in South Dakota, the state an individual hunter or trapper in 2013 compared passed a law that classifies wolves in the eastern half to 2012, and extended the wolf-killing season of the state as “varmints” that can be shot on sight. through the middle of March (when wolves would iii See: http://gazers.com/previously-endangered-wolves-can-now- iv Wisconsin Department of Natural Resources 2013 wolf exterminated-idaho/ hunting and trapping regulations; The Political Environment, April 12, 2013 8 IX. Conclusion In its rush to remove federal protections for gray wolves in most of the lower 48, the U.S. Fish and Wildlife Service relied on the states to adequately manage and conserve the species. As these examples above demonstrate, however, state management of wolves has been a political, rather than a science-based, endeavor. In the three years states have had wolf-management authority, nearly 3,000 wolves have been killed from hunting and trapping, sanctioned by state policies that fail to adequately consider the long-term viability or need for further recovery of wolves. The Service’s plan to now remove federal protections throughout most of the remaining lower 48 states and allow states to fully manage wolves not only jeopardizes the future of existing wolf populations it also makes it nearly impossible for dispersing wolves to make their way to adjacent states to establish new populations of wolves. To achieve true, long-term, sustainable, recovery of the gray wolf, federal wolf protections should be maintained and recovery plans developed, with the goal of restoring connected, resilient, ecologically-effective wolf populations wherever suitable wolf habitat exists. Formation of a recovery team made up of the many highly-qualified wolf biologists and other scientists could ensure that considerable recent science is brought to bear and ultimately produces a scientifically and legally defensible recovery strategy that specifies the conditions under which wolves are downlisted and ultimately delisted in all or portions of the species’ range. Restoring wolves to these areas would fulfill the ESA’s mandate to recover threatened or endangered species throughout all significant portions of their ranges and to conserve the ecosystems upon which they depend. Yellowstone wolf by doublejwebers / Flickr Bibliography 9 Bibliography Barber-Meyer, S.M., Mech, L.D. & White, P.J. 2008. Elk calf survival and mortality following wolf restoration to Yellowstone National Park. Wildlife Monographs, 169, 1–30. Berger, K.M., Gese, E.M., and J. Berger. 2008. Indirect effects and traditional trophic cascades: a test involving wolves, coyotes, and pronghorn. Ecology 89(3): 818–828. Carroll, Carlos, Phillips, M.K., Lopez-Gonzales, C.A. and Nathan H. Schumaker. 2006. Defining recovery goals and strategies for endangered species using spatially- explicit population models: the wolf as a case study. Bioscience 56:25-37. Eisenberg, C., Seager, S.T. and D.E. Hibbs. 2013. Wolf, elk, and aspen food web relationships: Context and complexity. Forest Ecol. Manage. (2013), http://dx.doi.org/10.1016/j.foreco.2013.01.014 Estes, J.A., Terborgh, J.A., Brashares, J.S., Power, M.E., Berger, J., Bond, W.J., Carpenter, S.R., Essington, T.E., Holt, R.D., Jackson, J.B.C., Marquis, R.J., Oksanen, L., Oksanen, T., Paine, R.T., Pikitich, E.K., Ripple, W.J., Sandin, S.A., Scheffer, M. 2011. Trophic downgrading of planet Earth. Science 33, 301–306. Frankel, O. H. and M. E. Soule. 1981. Conservation and Evolution. Cambridge University Press, New York, NY. Frankham R. 1995. Effective population size/adult population size ratios in wildlife: a review. Genetical Research 66: 95-107. Franklin, I. R. and R. Frankham. 1998. How large must a population be to retain evolutionary potential? Animal Conservation 1: 69-70. Hebblewhite, M., C.A. White, C.G. Nietvelt, J.A. McKenzie, T.E. Hurd, J.M. Fryxell, S.E. Bayley, P.C. Paquet. 2005. Human activity mediates a trophic cascade caused by wolves. Ecology 86(8):2135-2144. Hollenbeck, J., Ripple, W.J., 2008. Aspen patch and migratory bird relationships in the Northern Yellowstone Ecosystem. Landsc. Ecol. 22, 1411–1425. Leonard, J.A., C. Vila, and R.K. Wayne. 2005. Legacy lost: genetic variability and population size of extirpated U.S. grey wolves (Canis lupus). Molecular Ecology. 14(1): 9-17. Mech, L.D. and L. Boitani. Editors. 2003. Wolves: behavior, ecology and conservation. University of Chicago Press. Chicago, IL. Reed, D. H, J. J. O’Grady, B. W. Brook, J. D. Ballou and R. Frankham. 2003. Estimates of minimum viable population sizes for vertebrates and factors influencing those estimates. Biological Conservation 13:1 23-34. Ripple W.J. and R.L. Beschta. 2004. Wolves and the ecology of fear: can predation risk structure ecosystems? BioScience 54(8):755-766. Ripple W.J. and R.L. Beschta. 2012. Trophic Cascades in Yellowstone: The first 15 years after wolf reintroduction. Biological Conservation 145: 205–213 10 Ripple, W.J., Beschta, R.L., Fortin, J.K., and Robbins, C.T. 2013. Trophic Cascades from Wolves to Grizzly Bears in Yellowstone. Journal of Animal Ecology, doi: 10.1111/1365-2656.12123. Ripple, W.J., Estes, J.A., Beschta, R.L., Wilmers, C.C., Ritchie, E.G., Hebblewhite, M., Berger, J., Elmhagen, E., Letnic, M., Nelson, M.P., Schmitz, O.J., Smith, D.W., Wallach, A.D. and A.J. Wirsing. 2014. Status and ecological effects of the world’s largest carnivores. Science 343, 1241484. Seton, E.T. 1929. Lives of Game Animals, Vol. 1: Cats, wolves, and foxes. Doubleday and Doran Co., New York. Shaffer, M.L. 1981. Minimum population sizes for species conservation. BioScience 31: 131-134. Smith, Douglas W., R. 0. Peterson, and D. B. Houston. 2003. Yellowstone after wolves. Bioscience. 53(4):330-40. Soule, M. E. and B. A. Wilcox. (Eds.) 1980. Conservation Biology, an evolutionary-ecological perspective. Sinauer, Sunderland, MA. Soule, M. E. (1980). “Thresholds for survival: maintaining fitness and evolutionary potential.” Conservation biology: an evolutionary-ecological perspective, M. E. Soule and B. A. Wilcox, eds., Sinauer Associates, Sunderland, MA. 151-169. Soule, M. E. 1986. (Ed.) Conservation Biology, the science of scarcity and diversity. Sinauer, Sunderland, MA. Soule, M., Estes, J., Berger, J., and C. Del Rio. 2003. Ecological Effectiveness: Conservation Goals for Interactive Species. Conservation Biology, 17(5): 1238-1250. Stahler, D. R., D. W. Smith, and D. S. Guernsey. 2006. Foraging and feeding ecology of the gray wolf (Canis lupus): Lessons from Yellowstone National Park, Wyoming, USA. Journal of Nutrition. 136: 1,923s-1,926s. Traill, L. W., Corey J. A. Bradshaw, and Barry W. Brook. 2007. Minimum viable population size: a meta- analysis of 30 years of published estimates. Biological Conservation 139: 159-166. U.S. Fish and Wildlife Service. 1978. Reclassification of the Gray Wolf in the United States and Mexico, with Determination of Critical Habitat in Michigan and Minnesota. Federal Register, V. 43(47): 9607-9615. U.S. Fish and Wildlife Service. 2009. Final Rule to Identify the Western Great Lakes Populations of Gray Wolves as a Distinct Population Segment; Final Rule to Identify the Northern Rocky Mountain Population of Gray Wolf as a Distinct Population Segment; and To Revise the List of Endangered and Threatened Wildlife. Federal Register, V. 74(62): 15069-15123. U.S. Fish and Wildlife Service. 2011a. Reissuance of Final Rule To Identify the Northern Rocky Mountain Population of Gray Wolf as a Distinct Population Segment and To Revise the List of Endangered and Threatened Wildlife. Federal Register, V. 76(87): 25590-25592. U.S. Fish and Wildlife Service. 2011b. Final Rule Revising the Listing of the Gray Wolf (Canis lupus) in the Western Great Lakes. Federal Register, V. 76(249): 81666-81726. U.S. Fish and Wildlife Service. 2013. Proposed Rule Removing the Gray Wolf (Canis lupus) from the List of Threatened and Endangered Wildlife and Maintaining Protections for the Mexican Wolf (Canis lupus baileyi) by Listing it as Endangered; Proposed Revision to the Nonessential Experimental Population of the Mexican Wolf. Federal Register, V. 78(114): 35664-35719. 11 Weiss, A.E., Kroeger, T., Haney, J.C. and N. Fascione. 2007. Social and ecological benefits of restored wolf populations. Transactions of the 72nd North American Wildlife and Natural Resources Conference. Pp. 297- 319. Wild, M. A., M. W. Miller, and N. T. Hobbs. 2005. Could wolves control chronic wasting disease? Second International Chronic Wasting Disease Symposium. http://www.cwd-info.org/pdf/2005-cwd- symposiumprogram.pdf. Wilmers, C. C., and W. M. Getz. 2005. Gray wolves as climate change buffers in Yellowstone. PLoS Biology. 3(4):e92. Young, S.P. and E.A. Goldman. 1944. The Wolves of North America. American Wildlands Institute, Washington, D.C. Pp. 1-385. Federal Endangered Species Act. 16 U.S.C. §1531-1544. 12 APPENDIX A: Wolf Habitat Modeling Literature Reviewed Bednarz, J.C. 1989. An evaluation of the ecological potential of White Sands Missile Range to support a reintroduced population of Mexican wolves. Endangered Species Report No. 19. U.S. Fish and Wildlife Service, Albuquerque, NM. 96 pp. Belongie, Cole. 2008. Using GIS to Create a Gray Wolf Habitat Suitability Model and to Assess Wolf Pack Ranges in the Western Upper Peninsula of Michigan. Volume 10, Papers in Resource Analysis. 15pp. Saint Mary’s University of Minnesota Central Services Press. Winona, MN. Bennett, L.E. 1994. Colorado gray wolf recovery: A biological feasibility study. Final Report. U.S. Fish and Wildlife Service and University of Wyoming Fish and Wildlife Cooperative research unit, Laramie, Wyoming, USA. Available at: http://babel.hathitrust.org/cgi/pt?id=umn.31951p00672031a;view=1up;seq=146 Carroll, C. 2003. Wolf viability in the Northeastern United States and Southeastern Canada: A summary of new research with implications for wolf recovery. Richmond, VT: the Wildlands Project. Special Paper No. 5 Carroll, Carlos. 2005. Carnivore Restoration in the Northeastern U.S. and Southeastern Canada: A Regional- Scale Analysis of Habitat and Population Viability for Wolf, Lynx, and Marten (Report 2: Lynx and Marten Viability Analysis). Wildlands Project Special Paper No. 6. Richmond, VT: Wildlands Project. 46 pp. Carroll, C., Noss, R.F., Schumaker, N.H., and P.C. Paquet. 2001. Is the return of the wolf, wolverine, and grizzly bear to Oregon and California biologically feasible? In Large Mammal Restoration. Ecological and Sociological Challenges in the 21st Century. Maher, D.S., Noss, R.F. and J.L. Larkins (Eds.) Island Press. Washington, Covelo, London. Carroll, C., Phillips, M.K., Schumaker, N.H. and D.W. Smith. 2003. Impacts of Landscape Change on Wolf Restoration Success: Planning a Reintroduction Program Based on Static and Dynamic Spatial Models. Conservation Biology. Vol. 17(2): 536–548. Carroll, C. , Phillips, M.K. and C.A. Lopez-Gonzales. 2004. Spatial Analysis of Restoration Potential and Population Viability of the Wolf (Canis lupus) in the Southwestern United States and Northern Mexico, Prepared for the Turner Endangered Species Fund. Carroll, Carlos, Phillips, M.K., Lopez-Gonzales, C.A. and Nathan H. Schumaker. 2006. Defining recovery goals and strategies for endangered species using spatially- explicit population models: the wolf as a case study. Bioscience 56:25-37. Gehring, T.M. and B.A. Potter. 2005. Wolf Habitat Analysis in Michigan: An Example of the Need for Proactive Land Management for Carnivore Species. Wildlife Society Bulletin, Vol. 33, No. 4 (Winter, 2005), pp. 1237- 1244. Harrison, D.J. and T.B. Chapin. 1997. An Assessment of Potential Habitat for Eastern Timber Wolves in the Northeastern United States and Connectivity with Occupied Habitat in Southeastern Canada. A summary report and position paper prepared for the Wildlife Conservation Society. 13 Hearne, D., Lewis, K., Martin, M., Mitton, E. and C. Rocklen. 2003. Assessing the Landscape: Toward a Viable Gray Wolf Population in Michigan and Wisconsin. School of Natural Resources and Management, University of Michigan. Houts, M.E. 2001. Modeling Gray Wolf Habitat in the Northern Rocky Mountains. Department of Geography. University of Kansas. Houts, M.E. 2003. Using logistic regression to model wolf habitat suitability in the northern Rocky Mountains. World Wolf Congress, Banff, Alberta. Johnson, T.B., Noel, D.C. and Ward, L.Z. 1992. Summary of Information on Four Potential Mexican Wolf Reintroduction Areas in Arizona. Technical Report 23, Nongame and Endangered Wildlife Program, Arizona Game and Fish Department. October. Larsen T. and W.J. Ripple. 2006. Modeling gray wolf (Canis lupus) habitat in the Pacific Northwest. Journal of Conservation Planning. Vol 2: 17-33. Maletzke, B.T. and R.B. Wielgus. 2011. Development of wolf population models for RAMAS© analysis by the Washington Department of Fish and Wildlife. In Press. Mladenoff, D.J. and T.A. Sickley. 1998. Assessing potential gray wolf restoration in the northeastern United States: A spatial prediction of favorable habitat and potential population levels. Journal of Wildlife Management, V. 62: 1-10. Mladenoff, D.J., Sickley, T.A., Haight, R.G. and A.P. Wydevan. 1995. A Regional Landscape Analysis and Prediction of Favorable Gray Wolf Habitat in the Northern Great Lakes Region. Conservation Biology, Vol. 9, No. 2 (Apr., 1995), pp. 279-294. Mladenoff, D.J., Haight, R.G., Sickley, T.A. and A.P. Wydevan. 1997. Causes and Implications of Species Restoration in Altered Ecosystems. BioScience, Vol. 47, No. 1 (Jan., 1997), pp. 21-31. Mladenoff, D.J., Sickley, T.A. and A.P. Wydevan. 1999. Predicting Gray Wolf Landscape Recolonization: Logistic Regression Models Vs. New Field Data. Ecological Applications, 9(1), 1999, pp. 37– 44. Oakleaf, J.K., Murray, D.L., Oakleaf, J.R., Bangs, E.E., Mack, C.M., Smith, D.W., Fontaine, J.A., Jimenez, M.D., Meier, T.J. and C.C. Niemeyer. 2006. Habitat Selection by Recolonizing Wolves in the Northern Rocky Mountains of the United States. Journal of Wildlife Management 70(2):554-563. Paquet, P.C., J.R. Strittholt and N.L. Stauss. 1999. Wolf reintroduction feasibility in the Adirondack Park. Conservation Biology Institute, Corvallis, OR. Potvin, M.J., Drummer, T.D., Vucetich, J.A., Beyer, Jr., E. and J.H. Hammill. 2005. Monitoring and Habitat Analysis for Wolves in Upper Michigan. The Journal of Wildlife Management, Vol. 69, No. 4 (Oct., 2005), pp. 1660-1669. Ratti, J.T., Weinstein, M., Scott, J.M., Wiseman, P.A., Gillesberg, A.M., Miller, C.A., Szepanski, M.M., and Svencara, L.K. 2004. Feasibility of wolf reintroduction to Olympic Peninsula, Washington. Northwest Science 78:1-76. 14 Sneed, Paul G. 2001. The feasibility of gray wolf reintroduction to the Grand Canyon ecoregion. Endangered Species Update; Vol. 18(4). pp. 153-158. Wydevan, A.P., Fuller, T.K., Weber, W. and K. MacDonald. 1998. The Potential for Wolf Recovery in the Northeastern United States via Dispersal from Southeastern Canada. Wildlife Society Bulletin, Vol. 26, No. 4, Commemorative Issue Celebrating the 50th Anniversary of “A Sand County Almanac’’ and the Legacy of Aldo Leopold (Winter, 1998), pp.776-784. 15 APPENDIX B: Predicted Wolf Habitat Area and Wolf Populations # of Wolves Suitable Wolf Habitat Region Could Reference Study Region Method Study Location (mi2) Support Bednarz 1989 Southwest LANDSAT White Sands Missile Range, NM 996 32-40 Belongie 2008 Midwest GIS Overlay, Expert MI’s Upper Peninsula 4,846 - Ranking Bennett 1994 S. Rockies Unweighted Ranking CO 12,000 - 25,000 407 – 814 Carroll 2003 Northeast PATCH ME (2025) - 1,030 NH (2025) - 68 NY (2025) - 338 VT (2025) - 50 Carroll et al. 2001 West Coast GIS Index OR & CA 48,494 - Northeast OR 100 Southwest OR and Northern CA 190 – 470 Carroll et al. 2003 S. Rockies - - >1,000 Carroll et al. 2004 Southwest PATCH Blue Range AZ/NM, Grand Canyon, 17,390 - 24,131 3,166 AZ, Mogollon Rim, AZ, San Juans, CO, Vermejo/Carson, NM, UT Carroll et al. 2006 Western U.S. - 233,081 - Gehring and Potter 2005 Midwest Per Mladenoff et al. 1995 MI 849-1,634 40 – 105 Harrison and Chapin Northeast ARC Info and GIS ME 17,064 - 1997 NY 5,644 - NH 1,773 - VT 954 - MA 20 - ME & NH - 488 Adirondacks, NY - 146 Hearne et al. 2003 Midwest MI & WI - >1,000 Houts 2001 N. Rockies Logistic Regression Central ID, GYE, Western MT 27,751 - Houts 2003 West Coast Logistic Regression OR & WA 9,730 - N. Rockies Central ID, GYE, Western MT 117,000 - Johnson et al. 1992 Southwest AZ 14,099 - Larsen and Ripple 2006 N. Rockies Logistic Regression OR, ID, MT, WY 26,448 1,450 16 APPENDIX C: Predictive Modeling Parameters Used in 27 Wolf Habitat Suitability Models for the Lower 48 United States Reference Road density density Human or disturbance Land cover/use Land use/owner density Prey Prey accessability Prey availability Livestock density or Climate snowfall or Slope elevation / Soil depth hydrology water Surface availability Bednarz 1989 Belongie 2008 x x x x Bennett 1994 Carroll et al. 2001 x x x x x Carroll et al. 2003 x x x x Carroll et al. 2004 x x x x x Carroll et al. 2006 x x Carroll 2003 x x x Carroll 2005 x x x x x x Gerhing and Potter 2005 x Harrison and Chapin 1997 x x x Hearne et al. 2003 x x x Houts 2001 x x x x Houts 2003 x x Johnson et al. 1992 x x x x x x x Larsen and Ripple 2006 x x x x x x x Maletzke and Wielgus 2011 Mladenoff and Sickly 1998 x x Mladenoff et al. 1995 x x x x x Mladenoff et al. 1997 x x x x Mladenoff et al. 1999 x Oakleaf et al. 2006 x x x x x x x Paquet et al. 1999 x x x x x Potvin et al. 2005 x x Ratti et al. 2004 x x x x x x Sneed 2001 x x x x x x Wydevan et al. 1998 x x x 17 APPENDIX D: Tabulation of Dispersing Wolves, 1981-2014 Estimated Estimated Dispersal Reported End Origin Distance Date Location Location (mi) Outcome Source Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. Brown County, Shot and 3/15/81 Minnesota 139 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. South Dakota killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. Dickey County, Shot and 10/13/85 Minnesota 154 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. North Dakota killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. Harding County, Shot and 6/4/86 Unknown 349 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. South Dakota killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. Brown County, Shot and 11/10/89 Minnesota 158 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. South Dakota killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. McIntosh County, Shot and 2/27/90 Minnesota 185 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. North Dakota killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. Walsh County, Shot and 2/2/90 Minnesota 29 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. North Dakota killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. Mountrail County, Manitoba, Shot and 2/27/91 167 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. North Dakota Canada killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. Grant County, Shot and 12/1/91 Minnesota 126 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. South Dakota killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm 18 Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. Tripp County, Shot and 12/22/91 Minnesota 329 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. South Dakota killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm Licht, D.S. and S.H. Fritts. 1998. Gray Wolf (Canis lupus) occurrences in the Dakotas. U.S. Dunn County, Shot and 1/6/1992 Unknown 213 Fish and Wildlife Service. Jamestown, ND: Northern Prairie Wildlife Research Center Online. North Dakota killed Accessed at: http://www.npwrc.usgs.gov/resource/mammals/wolves/index.htm Maine Wolf Coalition. Wolves in the Northeast. Accessed at: Shot and http://mainewolfcoalition.org/wolves-in-the-northeast/; New York Times. 1996. Signs Suggest Aug-93 Northeast Maine Unknown Unknown killed1 a Return Of Timber Wolf to Maine. Accessed at: http://www.nytimes.com/1996/12/22/us/signs-suggest-a-return-of-timber-wolf-to-maine.html Boyd County, Western Great Shot and U.S. Fish and Wildlife Service. 2011. Revisiting the Listing of the Gray Wolf in the Western 1994/95 300 Nebraska Lakes DPS killed Great Lakes. Federal Register, Vol. 76, No. 249. Maine Wolf Coalition. Wolves in the Northeast. Accessed at: Shot and http://mainewolfcoalition.org/wolves-in-the-northeast/; New York Times. 1996. Signs Suggest Nov-96 Eastern Maine Unknown Unknown killed a Return Of Timber Wolf to Maine. Accessed at: http://www.nytimes.com/1996/12/22/us/signs-suggest-a-return-of-timber-wolf-to-maine.html Maine Wolf Coalition. Wolves in the Northeast. Accessed at: Shot and http://mainewolfcoalition.org/wolves-in-the-northeast/; Urbigkit, C. 2007. Wolf killed in Nov-98 Glover, Vermont Unknown Unknown killed Vermont. Pinedale Online! Accessed at: http://www.pinedaleonline.com/news/2007/10/WolfkilledinVermont.htm Harding County, Western Great U.S. Fish and Wildlife Service. 2011. Revisiting the Listing of the Gray Wolf in the Western 2001 400 Killed South Dakota Lakes DPS Great Lakes. Federal Register, Vol. 76, No. 249. Maine Wolf Coalition. Wolves in the Northeast. Accessed at: http://mainewolfcoalition.org/wolves-in-the-northeast/; Nearing, B. 2011. A century later, the Shot and Dec-01 Day, New York Unknown Unknown wild wolf returns: Animal shot a decade ago hints at the arrival of predator in New York. killed (MC) Times Union. Accessed at: http://www.timesunion.com/local/article/A-century-later-the-wild- wolf-returns-2222681.php Pend Oreille Northwestern Moved into Wiles, G.J., H.L. Allen, and G.E. Hayes. 2011. Wolf conservation and management plan for Feb-02 County, Unknown Montana B.C. Washington. Washington Department of Fish and Wildlife, Olympia, Washington. 297 pp. Washington 19 Michigan Department of Natural Resources. 2008. Michigan Wolf Management Plan. Wildlife Ironwood, Outside of Trenton, Division Report No. 3484. Accessed at: Gogebic Shot and 10/23/02 Grundy County, 470 http://www.michigan.gov/documents/dnr/Draft_Wolf_Management_Plan_030708_227742_7. County, killed Missouri pdf; Beringer, J. Personal communication with Jeff Beringer, Resource Scientist, Missouri Michigan Department of Conservation. May 9, 2013. University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: Henry, Marshall Shot and http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Kath, J.A. 2013. Dec-02 Wisconsin 200 County, Illinois killed (MC) Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois Department of Natural Resources. May 15, 2013. U.S. Fish and Wildlife Service. 2003. Status of Gray Wolf Recovery, Weeks of 3/28 to 4/04, Spalding, Greeley Western Great Shot and 12/15/02 350 2003. Weekly Report, Gray Wolves in the Northern Rocky Mountains. Accessed at: County, Nebraska Lakes DPS killed http://www.fws.gov/mountain-prairie/species/mammals/wolf/WeeklyRpt03/wk04042003.htm Trapped and Yellowstone Hollenhorst, J. 2002. Wolf Capture in Utah Ignites Controversy. KSL-TV. Accessed at: 2002 Morgan, Utah 240 released in NP http://web.ksl.com/dump/news/cc/local/wolf_return.php Yellowstone Southwest (SW 1/4 Wisconsin Department of Natural Resources. 1999. Wisconsin Wolf Management Plan. Jackson Sec. 10, T19N, Madison, Wisconsin. Accessed at: http://dnr.wi.gov/files/pdf/pubs/er/er0099.pdf; 2003 County, 428 Found dead R13E) Randolph http://www.predatormastersforums.com/forums/ubbthreads.php?ubb=showflat&Number=8926 Wisconsin County, Indiana 1 Northwest U.S. Fish and Wildlife Service. 2004. News Release: Preliminary Necropsy Results For Gray I-70, 30 mi west of corner of Killed by 6/7/04 446 Wolf Found Dead Near Denver, Colorado. U.S. Fish and Wildlife Mountain-Prairie Region, Denver, Colorado Yellowstone motor vehicle Lakewood, Colorado. Accessed at: http://www.fws.gov/mountain-prairie/pressrel/04-43.htm NP Upper Lower Peninsula, U.S. Fish and Wildlife Service. 2011. Revisiting the Listing of the Gray Wolf in the Western Oct-04 Peninsula, Unknown Killed Michigan Great Lakes. Federal Register, Vol. 76, No. 249. Michigan University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: Chain O'Lakes Killed by http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Kath, J.A. 2013. 2/17/05 State Park, Lake Wisconsin Unknown motor vehicle Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois County, Illinois Department of Natural Resources. May 15, 2013. U.S. Fish and Wildlife Service. 2011. Revisiting the Listing of the Gray Wolf in the Western Great Lakes. Federal Register, Vol. 76, No. 249; University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: New Canton, Pike Central Shot and http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Young, C. 2006. Dec-05 300 County, Illinois Wisconsin killed Wild wolf killed in Pike County Illinois. Wolf Saga. Accessed at: http://wolfsaga.blogspot.com/2006/02/wild-wolf-killed-in-pike-county.html; Kath, J.A. 2013. Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois Department of Natural Resources. May 15, 2013. 20 Maughan, R. 2006. Wolf hit on I-90 near Sturgis, SD last April was a Yellowstone area wolf. Sturgis, South Yellowstone Killed by Wolf Report 8-15-2006. Accessed at: http://www.forwolves.org/ralph/sturgis-wolf.htm; Stark, Apr-06 300 Dakota NP motor vehicle M. 2007. Yellowstone wolf hit near Sturgis. The Billings Gazette. Accessed at: http://helenair.com/news/article_cee1229e-ec91-575c-a7ea-30287aac9b33.html Maine Wolf Coalition. Wolves in the Northeast. Accessed at: North Troy, Shot and http://mainewolfcoalition.org/wolves-in-the-northeast/; Urbigkit, C. 2007. Wolf killed in Oct-06 Unknown Unknown Vermont killed Vermont. Pinedale Online! Accessed at: http://www.pinedaleonline.com/news/2007/10/WolfkilledinVermont.htm Northern U.S. Fish and Wildlife Service. 2007. Rocky Mountain Wolf Recovery 2006 Annual Report. 2006 Tremonton, Utah Rocky Unknown Killed C.A. Sime and E.E. Gangs, eds. USFWS, Ecological Services, 585 Shepard Way, Helena, Mountain DPS Montana. 59601. 235 pp. Maine Wolf Coalition. Wolves in the Northeast. Accessed at: Shelburne, Shot and http://mainewolfcoalition.org/wolves-in-the-northeast/; Reitz, S. 2008. Rare Gray Wolf Oct-07 Unknown Unknown Massachusetts killed Appears in Western Massachusetts. National Geographic News. Accessed at: http://news.nationalgeographic.com/news/2008/03/080305-AP-wolf-return.html University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: Western Great Shot and http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Kath, J.A. 2013. 2/19/08 Lena, Illinois Unknown Lakes DPS killed (MC) Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois Department of Natural Resources. May 15, 2013. "Coastal B.C. and NE B.C., NW Alberta, or Okanogan County, Wiles, G.J., H.L. Allen, and G.E. Hayes. 2011. Wolf conservation and management plan for 2008 reintroduced Unknown Unknown Washington Washington. Washington Department of Fish and Wildlife, Olympia, Washington. 297 pp. populations in central ID and GYE” Coastal/Southe U.S. Fish and Wildlife Service. 2011. Rocky Mountain Wolf Recovery 2010 Interagency Jul-08 Twisp, Washington rn British Unknown Unknown Annual Report. C.A. Sime and E.E. Bangs, eds. USFWS, Ecological Services, 585 Shepard Columbia Way, Helena, Montana, 59601. University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: Kane County, Shot and http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Kath, J.A. 2013. 2009 Unknown Unknown Illinois killed (MC) Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois Department of Natural Resources. May 15, 2013. South of Killed by The Southwest Environmental Center. 2011. Lone wolf travels 3,000 miles before being South of Meeker, Apr-09 Bozeman, 400 Compound poisoned. Accessed at: http://www.wildmesquite.org/news/lone-wolf-travels-3000-miles- Colorado Montana 1080 being-poisoned/071511 21 Pend Oreille Northwestern Wiles, G.J., H.L. Allen, and G.E. Hayes. 2011. Wolf conservation and management plan for May-09 County, Montana/north Unknown Unknown Washington. Washington Department of Fish and Wildlife, Olympia, Washington. 297 pp. Washington ern Idaho Bellevue, Butterworth, S. 2013. Personal communication with Scott Butterworth, District Manager, Shot and Mar-10 Sandusky County, Unknown Unknown Wildlife District Two, Ohio Department of Natural Resources. May 14, 2013.; killed Ohio http://www.toledoblade.com/StevePollick/2010/03/19/Wolf-shooting-piques-curiosity.html Shot and Linnell, M. 2013. Personal communication with Mike Linnell, State Director, Utah, USDA- Jul-10 Rich County, Utah Unknown Unknown killed Wildlife Services. May 10, 2013. Northern Shot and Cache County, Love, C. 2010. Wolf Range Expands into Utah. Field and Stream. Accessed at: 2010 Rocky Unknown killed in Utah http://www.fieldandstream.com/blogs/hunting/2010/07/wolf-range-expands-utah Mountain DPS Idaho 12 miles northeast Jerek, J. 2013. DNA shows hunter-shot canine from October to be wandering wolf. Missouri of Carrolton, Western Great Shot and Department of Conservation. Accessed at: http://mdc.mo.gov/newsroom/dna-shows-hunter- 2010 Unknown Carroll County, Lakes DPS killed (MC) shot-canine-october-be-wandering-wolf; Beringer, J. Personal communication with Jeff Missouri Beringer, Resource Scientist, Missouri Department of Conservation. May 9, 2013. Two packs Northern U.S. Fish and Wildlife Service. 2012. Northern Rocky Mountain Wolf Recovery Program Western established 2011 Rocky Unknown 2011 Interagency Annual Report. M.D. Jimenez and S.A. Becker, eds. USFWS, Ecological Washington outside of Mountain DPS Services, 585 Shepard Way, Helena, Montana, 59601. NRM DPS University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: Southeast Jo Shot and http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Kath, J.A. 2013. 2011 Daviess County, Unknown Unknown killed (male) Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois Illinois Department of Natural Resources. May 15, 2013. University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: Southeast Jo Shot and http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Kath, J.A. 2013. 2011 Daviess County, Unknown Unknown killed Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois Illinois (female) Department of Natural Resources. May 15, 2013. Hillsboro, North Shot and Associated Press. 2011. Federally protected gray wolf shot in North Dakota. Twin Cities Jan-11 Unknown Unknown Dakota killed (MC) Pioneer Press. Accessed at: http://www.twincities.com/news/ci_17555251 Beringer, J. Personal communication with Jeff Beringer, Resource Scientist, Missouri Clinton County, Western Great Shot and Department of Conservation. May 9, 2013; Leonard, J. 2012. Gower man unknowingly bags Oct-11 Unknown Missouri Lakes DPS killed (MC) wolf. News-Press Now. Accessed at: http://www.newspressnow.com/sports/outdoors/ article_917229af-fbd3-5ba4-9386-05f9e928cec9.html 22 Travelled Oregon Department of Fish and Wildlife. 2013. Wolves in Oregon: Wolf Program Updates. Northeastern back to Accessed at: http://www.dfw.state.or.us/wolves/; California Department of Fish and Wildlife. 12/28/11 Northern California Oregon 350 Oregon in Wildlife and Habitat Management: OR-7 - A Lone Wolf's Story. Accessed at: (Imnaha pack) March 2013 http://www.dfg.ca.gov/wildlife/nongame/wolf/OR7story.html Woster, K. 2012. Wolf killed near Custer likely from Great Lakes population. Rapid City Journal. Accessed at: http://rapidcityjournal.com/news/wolf-killed-near-custer-likely-from- Custer County, Western Great Feb-12 Unknown Killed great-lakes-population/article_92dc1eb8-6be8-11e1-8ce3-001871e3ce6c.html; Custer County South Dakota Lakes DPS Chronicle Online. 2012. Wolf killed near Custer. Accessed at: http://www.custercountynews.com/cms/news/story-427443.html Southeast Found dead, Woster, K. 2012. Wolf found near Pine Ridge migrated from Yellowstone. Rapid City Journal. Pine Ridge, South May-12 Yellowstone 400 likely hit by Accessed at: http://rapidcityjournal.com/news/wolf-found-near-pine-ridge-migrated-from- Dakota NP motor vehicle yellowstone/article_f6d01210-9f07-11e1-a76d-001a4bcf887a.html Franklin Island Jerek, J. 2013. DNA shows hunter-shot canine from October to be wandering wolf. Missouri Conservation Area, Western Great Shot and Department of Conservation. Accessed at: http://mdc.mo.gov/newsroom/dna-shows-hunter- Oct-12 Unknown Howard County, Lakes DPS killed (MC) shot-canine-october-be-wandering-wolf; Beringer, J. Personal communication with Jeff Missouri Beringer, Resource Scientist, Missouri Department of Conservation. May 9, 2013. Kansas Department of Wildlife, Parks and Tourism. 2013. Weekly News: Wolf Found In Trego County, Western Great Shot and Kansas. Accessed at: http://kdwpt.state.ks.us/KDWPT-Info/News/Weekly-News/1-31- Dec-12 Unknown Kansas Lakes DPS killed (MC) 13/WOLF-FOUND-IN-KANSAS; Peek, M. 2013. Personal Communication with Matt Peek, Wildlife Research Biologist, Kansas Department of Wildlife, Parks & Tourism. May 10, 2013. University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: Coleta, Whiteside Trapped and http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Kath, J.A. 2013. Dec-12 Unknown Unknown County, Illinois released Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois Department of Natural Resources. May 15, 2013. University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: Apple Canyon http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Kath, J.A. 2013. Mar-13 Lake, Jo Daviess Wisconsin 250 Found dead Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois County, Illinois Department of Natural Resources. May 15, 2013. University of Illinois. 2013. Wildlife Directory: Gray Wolf (Canis lupus). Accessed at: La Salle County, Killed by http://web.extension.illinois.edu/wildlife/directory_show.cfm?species=wolf; Kath, J.A. 2013. Dec-13 Unknown Unknown IL vehicle Personal Communication with Joseph A. Kath, Endangered Species Manager, Illinois Department of Natural Resources. May 15, 2013. Munfordville, Associated Press. 2013. Officials confirm gray wolf killed in Ky. Accessed at: Western Great Shot and Mar-14 Kentucky (Hart 750 http://www.wkyt.com/home/headlines/officials-confirm-gray-wolf-killed-in-ky- Lakes DPS killed County) 219749481.html 23 Love, Orlan. 2014. First wolf in Iowa in 89 years killed in Buchanan County. Accessed at: Fairbank, Iowa Western Great Shot and Feb-14 250 http://www.kcrg.com/subject/news/first-wolf-in-iowa-in-89-years-killed-in-buchanan-county- (Buchanan County) Lakes DPS killed (MC) 20140506 Awaiting info Shot and Love, Orlan. 2014. http://thegazette.com/subject/environment/nature/endangered- May-14 Jones County, Iowa from Iowa Unknown killed (MC) species/another-wolf-slain-in-iowa-20140717 DNR MC – Wolf was mistaken for coyote when shot. 1 shot by bear hunter who was prosecuted by USFWS 24 JOURNAL OF WILDLIFE DISEASES THE ROLE OF PREDATION IN DISEASE CONTROL: A COMPARISON OF SELECTIVE AND NONSELECTIVE REMOVAL ON PRION DISEASE DYNAMICS IN DEER Margaret A. Wild,1,5 N. Thompson Hobbs,2 Mark S. Graham,1,4 and Michael W. Miller3 1 National Park Service, Biological Resource Management Division, 1201 Oak Ridge Drive, Suite 200, Fort Collins, Colorado 80525, USA 2 Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, Colorado 80523-1499, USA 3 Colorado Division of Wildlife, Wildlife Research Center, 317 West Prospect Road, Fort Collins, Colorado 80526-2097, USA 4 Current address: National Park Service, New River Gorge National River, PO Box 246, 104 Main Street, Glen Jean, West Virginia 25846, USA Journal of Wildlife Diseases, 47(1), 2011, pp. 78–93 # Wildlife Disease Association 2011 THE ROLE OF PREDATION IN DISEASE CONTROL: A COMPARISON OF SELECTIVE AND NONSELECTIVE REMOVAL ON PRION DISEASE DYNAMICS IN DEER Margaret A. Wild,1,5 N. Thompson Hobbs,2 Mark S. Graham,1,4 and Michael W. Miller3 1 National Park Service, Biological Resource Management Division, 1201 Oak Ridge Drive, Suite 200, Fort Collins, Colorado 80525, USA 2 Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, Colorado 80523-1499, USA 3 Colorado Division of Wildlife, Wildlife Research Center, 317 West Prospect Road, Fort Collins, Colorado 80526-2097, USA 4 Current address: National Park Service, New River Gorge National River, PO Box 246, 104 Main Street, Glen Jean, West Virginia 25846, USA 5 Corresponding author (email: [email protected]) ABSTRACT: Effective measures for controlling chronic wasting disease (CWD), a contagious prion disease of cervids, remain elusive. We review theoretic relationships between predation and host- parasite dynamics and describe a mathematical model to evaluate the potential influence of random removal through harvest or culling and selective predation by wolves (Canis lupus) upon CWD dynamics in deer (Odocoileus spp.) populations. Imposing nonselective mortality representing a 15% annual harvest or cull 51 yr after CWD introduction lowered both deer population size and steady state CWD. Selective (43) mortality at the same 15% predation rate caused a more modest reduction in deer population size accompanied by a relatively rapid decline in CWD prevalence and elimination of the disease from a closed population. The impacts of selective predation on epidemic dynamics were sensitive to assumptions on parameter estimates; however, within expected ranges, the results of selective predation were consistent and robust. We suggest that as CWD distribution and wolf range overlap in the future, wolf predation may suppress disease emergence or limit prevalence. Key words: Canis lupus, chronic wasting disease, deer, host-parasite, Odocoileus spp., predator-prey, selective predation, wolf. INTRODUCTION et al., 2005; Johnson et al., 2007; Pedersen et al., 2007). Disease emergence and reemergence Changes in predation rates or predator- threaten the abundance and viability of prey dynamics are among the factors that wildlife species worldwide (Daszak et al., may affect patterns of disease emergence, 2000). Although a variety of factors appear reemergence, and persistence (Choo et to be contributing to the recent surges in al., 2003; Packer et al., 2003; Holt and diseases impacting natural populations, Roy, 2007). The potential effects of ecosystems altered by human activities predation on epidemic dynamics vary seem particularly vulnerable to such depending on both the nature of predation effects (Harvell et al., 1999; Daszak et occurring upon hosts and attributes of the al., 2000, 2001; Kutz et al., 2005; Johnson host-parasite relationship. Nonselective et al., 2007; Pedersen et al., 2007). predation could dampen epidemic dynam- Ecologic imbalances can diminish the ics by reducing host densities and contact resilience of host species to natural rates or by lowering the total number of fluctuations in pathogens and the host’s infected individuals in a host population capacity to resist or recover from pathogen (Heesterbeek and Roberts, 1995; Barlow, introductions. Such impacts on resiliency 1996; Packer et al., 2003). Similarly, can be observed with alterations to host- selective predation on infected individuals parasite relationships resulting in changes could eliminate pathogens or prevent their in host survival and contact rates among establishment under some circumstances susceptible and infected individuals (Har- (Heesterbeek and Roberts, 1995; Gross vell et al., 1999; Daszak et al., 2001; Kutz and Miller, 2001; Packer et al., 2003). 78 WILD ET AL.—PREDATION IN DISEASE CONTROL 79 Alternatively, both nonselective and selec- individual. Second, the effect of reduced tive predation might facilitate pathogen intervals of infectivity is amplified by emergence and persistence in cases where reductions in population density that occur resistant individuals become less abundant as mortality increases; such reductions (Choisy and Rohani, 2006; Holt and Roy, cause declines in the number of contacts 2007) and in cases where infected indi- between infected and susceptible individu- viduals are avoided by predators (Packer als. Both of these mechanisms slow rates of et al., 2003). It follows that ecosystems transmission of disease. If these mecha- altered by removal of natural predators by nisms cause the number of new infections humans may respond differently to en- produced per infected individual to fall demic or novel pathogens than intact below one, then the disease will be systems. eliminated from the population. In light of the potential influence of predation on host-parasite dynamics, the Selective predation: Any elevation in mortal- role of predators should be considered in ity rate has the potential to cause the devising strategies for control of emerging foregoing effects. Reductions in transmis- or reemerging pathogens in natural popu- sion rates and disease prevalence can be lations. We review the theoretic relation- particularly large if mortality rates are ships between predation and host-parasite disproportionately higher in the infected dynamics, using the term parasite broadly portion of the population than in the to describe any infectious agent capable of susceptible portion (Heesterbeek and Rob- infecting a host, utilizing host resources, erts, 1995). This explains why diseases that and spreading to new hosts (Altizer et al., cause rapid death fail to persist. However, 2003). We then describe a simple mathe- other, nondisease, agents of selective mor- matical model developed to evaluate how tality can exert the same beneficial effect. dynamics of prion disease in deer (Odocoi- For example, if predators prey selectively leus spp.) populations may respond to on diseased individuals, it is reasonable to nonrandom removal resulting from selec- expect that they might reduce disease tive predation by wolves (Canis lupus)and prevalence much more rapidly than would compare this outcome with effects of occur if mortality were nonselective. random removal through harvest or culling. Evidence that predators have a greater selectivity for diseased prey has been widely Predation and host-parasite dynamics observed. Vorˇı´sˇek et al. (1998) found Nonselective predation: The interplay among parasitized voles in buzzards’ diets in a host regulation, immune response, and the greater proportion than they occurred in pattern of predator selectivity determines the population. Birds with high blood whether predation reduces or increases the parasite loads (Moller and Nielsen, 2007) prevalence of disease in a population (Holt and birds with weakened immune systems and Roy, 2007). Under many scenarios, (Moller and Erritzoe, 2000) were preyed increasing mortality rates in diseased pop- upon at higher rates than uncompromised ulations can retard disease transmission and birds. Murray et al. (1997) reported in- reduce disease prevalence (Barlow, 1996; creased predation on snowshoe hares (Le- Lafferty and Holt, 2003; Packer et al., 2003; pus americanus) with heavy burdens of the Ostfeld and Holt, 2004). Increasing mortal- sublethal nematode Obeliscoides cuniculi ity slows transmission via two mechanisms. during periods of limited food supplies. First, it reduces the average lifetime of It is logical to assume that predators’ infected individuals. Reduced lifespan, in high success with diseased prey may be turn, can truncate the time interval when due to poorer body condition of the prey animals are infectious, thereby reducing the and consequently prey’s slower avoidance number of infections produced per infected behavior, decreased awareness, or re- 80 JOURNAL OF WILDLIFE DISEASES, VOL. 47, NO. 1, JANUARY 2011 duced stamina. Studies have suggested American cervid species (Spraker et al., that predators may also use visual pattern, 1997; Baeten et al., 2007). The origins and scent, or behavioral cues to select com- evolutionary history of CWD are unclear, promised prey. Hudson et al. (1992) but uncontrolled epidemics have the suggested that heavily parasitized female potential to depress deer populations red grouse (Lagopus lagopus scoticus) (Williams and Young, 1992; Miller et al., emitted more scent, and were, as a result, 2000, 2006; Gross and Miller, 2001; more easily detected by mammalian pred- Williams et al., 2002) and to impact ators. Larks (Calandrella rufescens) that ecosystems dominated by these species were infected with poxvirus had shorter, (Hobbs, 1996). Epidemics of CWD are lower-pitched distress calls than uninfect- sustained naturally by horizontal transmis- ed birds, indicating a behavioral change sion (Miller and Williams, 2003; Miller et that could affect predation rates (Laiolo et al., 2006), with both infected animals and al., 2007). Lafferty and Morris (1996) contaminated environments serving as reported that parasitized killifish (Fundu- sources of infection (Miller and Williams, lus parvipinnis) exhibited more conspicu- 2003; Miller et al., 2004, 2006; Mathiason ous behavior than uninfected killifish, and et al., 2006, 2009; Tamgu¨ney et al., 2009). were also preyed upon more heavily by Under some conditions, the CWD agent birds. Red-legged frog (Rana aurora) persists in the environment for years in tadpoles also exhibited modified behavior residues from excrement and infected when infected with yeast (Candida humi- carcasses (Miller et al., 2004). Mecha- cola), resulting in changes in thermoreg- nisms for both direct (animal-animal) and ulatory behavior, compromised predator indirect (animal-environment-animal) pri- avoidance behavior, and increases in being on transmission have been demonstrated preyed upon (Lefcort and Blaustein, empirically (Miller et al., 2004; Mathiason 1995). Examples of increased vulnerability et al., 2006, 2009; Tamgu¨ney et al., 2009), to selective predation in large mammals but models incorporating indirect trans- are less numerous; however, diseased mission best represent epidemic dynamics moose (Alces alces;JolyandMessier, in captive deer (Miller et al., 2006). 2004a) and bison (Bison bison; Joly and Effective measures for controlling Messier, 2004b) appeared to be more CWD remain elusive. In the absence of susceptible to predation by wolves than vaccines or therapies, strategies undertak- apparently healthy animals. White-tailed en to combat CWD have focused on deer (Odocoileus virginianus) killed by depressing the abundance of host species wolves may appear normal to human either locally or regionally in an attempt to inspection, but subtle alterations may be disrupt prion transmission (Williams et al., present as demonstrated by the correla- 2002; Grear et al., 2006; Conner et al., tion of fawn and subadult survival to 2007). Thus far, control strategies relying maternal and grand-maternal nutrition on hunting or culling by humans to lower (Mech et al., 1991). Further, Krumm et deer numbers and subsequently CWD al. (2009) recently reported that mountain prevalence have not yielded demonstrable lions (Puma concolor) prey selectively on effects (Conner et al., 2007). However, prion-infected mule deer (Odocoileus these results are not surprising given the hemionus) in Colorado, USA. limited duration of such management actions and because theory suggests that Wolves, selective predation, and prion randomly removing individuals from an disease dynamics infected population should have less effect Chronic wasting disease (CWD; Wil- on epidemic dynamics than selectively liams and Young, 1980) is a contagious removing infected individuals (Heester- prion disease of at least four North beek and Roberts, 1995; Gross and Miller, WILD ET AL.—PREDATION IN DISEASE CONTROL 81 2001). The protracted course of CWD in MATERIALS AND METHODS deer (Williams and Young, 1980, 1992; Model structure Fox et al., 2006) and occurrence of agent shedding well before the hallmark signs of We explored the potential impacts of emaciation and behavioral changes are predation on dynamics of deer populations using a simple model of interactions among discernable to human observers (Mathia- infected animals, susceptible animals, and son et al., 2009; Tamgu¨ney et al., 2009) infectious residue in the environment in a suggest that selectively removing only closed population. We derived the model used obviously ill deer from a population would here (Appendix A) from the indirect transmis- not be an effective control strategy (Gross sion model of Miller et al. (2006), which was the best approximating model of two CWD and Miller, 2001). If infected deer were epidemics in a captive population of mule detectable earlier in the disease course, deer. Because of the similarities in CWD however, selective removal might be more epidemiology between mule deer and white- effective than random removal in control- tailed deer (Miller and Wild, 2004), here we ling epidemics (Gross and Miller, 2001; generalize inferences to ‘‘deer.’’ We modified the best approximating model found by Miller Wolfe et al., 2004). et al. (2006) to portray disease dynamics in Increased vulnerability of CWD-infect- free-ranging populations as follows: ed mule deer to vehicle collisions (Krumm 1) We assumed that transmission rates were et al., 2005) suggests that lowered vigi- approximately 25 times lower in natural lance also might make them more vulner- populations than in captive ones. This able to large predators. It follows that if assumption was based on the elevated natural predators were able to develop a densities of deer in captive populations search image for subtle behavioral chang- (Miller et al., 2006). Adjusting transmission rates for differences in density was plausi- es of CWD infection in deer, then ble; however, the magnitude of the adjust- fostering predation upon CWD-infected ment for transmission was uncertain. deer populations might offer a viable Therefore we targeted this adjustment as adjunct or alternative to other control one of the variables to be explored in measures. Although mountain lions do simulation studies through the use of a scaling coefficient. appear to preferentially prey on mule deer 2) Per-capita birth rates were assumed to infected with CWD (Krumm et al., 2009), decline linearly with increasing population epidemics persist in mule deer herds in density. the presence of mountain lion predation 3) We added a term representing predation. (Miller et al., 2008; Krumm et al., 2009). This term could be adjusted to reflect Based on the subtlety of the behavioral selective predation, where predators favored infected animals over susceptible ones, or changes early in the course of CWD nonselective predation, as would occur with infection, we would expect coursing pred- hunting or culling, where removals were ators like wolves to show even greater assumed to be random. In the case of potential selective capability than ambush selective predation on diseased animals, we predators like mountain lions; however, also included a term to represent the extent to which predation mortality was compensatory wolves were extirpated and packs are with CWD mortality. presently absent from the areas in North America where CWD is endemic in deer, We sought to use the simplest model possible to achieve the greatest generality of results so field data are not available for compar- (Levins, 1966) and to reduce the number of ison. Consequently, to assess this possibil- parameters that had to be estimated. We ity we developed and explored the behav- avoided the use of an age-structured model, ior of models representing the effects of which would have required estimating unknown selective predation by wolves and com- transmission rates for several age classes. Dynamics of the prey population was not pared these with nonselective predation, coupled to the dynamics of predators and such as through harvest or culling, on predation intensity did not change with prey CWD dynamics in deer. abundance. In the interest of parsimony, we 82 JOURNAL OF WILDLIFE DISEASES, VOL. 47, NO. 1, JANUARY 2011 used a constant relative rate for predation. animals. Adding a single infected deer in Preliminary modeling included a type II func- year 1 produced oscillatory dynamics typical tional response and did not yield results that of epidemics. With disease and no predation, were qualitatively different than those present- ed here. More importantly, by holding preda- the equilibrium density was 736 deer and tion constant, we could be sure that observed disease prevalence was 29%.Thus,the dynamics resulted from the interplay between disease reduced animal abundance in our CWD and deer, rather than between deer and model by almost a third (Fig. 1A, B). Our wolves (analogous to choosing to hold one factor model resembles classic susceptible-infected constant in a designed experiment). Although our model is simple, we believe it represents the (SI) models with an additional mortality essential interactions in the deer-CWD system. source from predation and an environmental As knowledge of parameters improves, more reservoir of infection. Models of this general detailed models will be justified. type are known to have conditions that allow steady states (Miller et al., 2006), and the Model experiments model used here shows that equilibrium. We exercised the model to examine how Simulated selective and nonselective pre- selective and nonselective predation may influ- dation affected epidemic dynamics to differ- ence CWD prevalence. We made three model runs using our most plausible estimates of ent degrees. Imposing nonselective mortality model parameters to examine differences representing a 15% annual harvest or cull in among trajectories of diseased populations in year 51 lowered both deer population size the presence and absence of predation. We first and steady state CWD prevalence; however, conducted a reference simulation introducing a under the assumptions of this simulation, the single infected animal into a population of 1,000 deer at time50 and allowed the model to disease was able to persist in the population equilibrate over 100 yr. In two experimental (Fig. 1C, D). Selective (43) mortality at the simulations, we introduced predation in year 51. same 15% predation rate beginning in year In one of these simulations, predation was 51 caused a more modest reduction in deer assumed to occur randomly; in the other, population size accompanied by a relatively predators were assumed to favor infected rapid decline in CWD prevalence and individuals. In the case of selective predation, the modeled 15% predation rate was equivalent elimination of the disease from this closed to about seven wolves removing 16 deer/wolf/yr population (Fig. 1E, F). The impacts of (Mech and Peterson, 2003). selective predation on epidemic dynamics Although most parameters in the model were were sensitive to assumptions on vulnerabil- derived from Miller et al. (2006) or from ity of infected animals and compensation reasonable assumptions on deer biology (See Table 1 in Appendix A for all parameter values), between predation and mortality due to there was substantial uncertainty in our best CWD, as well as overall predation and CWD guesses of the value of several parameters transmission rates (Fig. 2). Doubling the controlling the effects of the disease and of vulnerability of infected animals to selective predation. Notable among these were the extent predation accelerated the rate of decline in of predator selectivity for CWD-infected ani- mals, the extent of compensation between prevalence (Fig. 2A). Increasing the propor- CWD and predator mortality, and the adjust- tion of compensatory deaths among infected ment for the rate of transmission in free-ranging deer dampened the predicted decline in populations. We explored consequences of prevalence (Fig. 2B); when compensation these uncertainties by conducting model exper- exceeded ca. 60%, selective predation had iments varying these parameters singly and in less of a predicted effect on epidemic pairs to examine the sensitivity of model predictions to uncertainty in their estimates. dynamics than nonselective predation. The overall predation rate also affected the rate RESULTS and magnitude of decline in steady state prevalence (Fig. 2C). Epidemic dynamics In the absence of CWD and predation, also were sensitive to assumptions on values the modeled deer population stabilized at an for the scaling of transmission rate (Fig. 2D), ecologic carrying capacity of about 1,000 with asymptotic prevalence varying by a WILD ET AL.—PREDATION IN DISEASE CONTROL 83 FIGURE 1. Simulations of deer abundance and disease prevalence in populations infected with CWD assuming no predation (A, B), nonselective predation (C, D), and selective predation (E, F). Lines in population-number graphs are number of susceptible deer (solid), number of infected deer (dashed), and total population (dotted). In the absence of CWD and predation, the population would reach equilibrium at 1,000 animals. We assumed that predators consumed four times more infected animals than would be expected by random selection among susceptible and infected deer and that compensation of predation for disease is 0.3. The predation rate was set at 0.15 beginning in year 51 and the scaling factor for transmission rate was 25 in all simulations. factor of more than 10 when changing the expected to greatly diminish the persis- scaling coefficient from 20 to 40. tence of CWD in a susceptible deer Our models predicted that interactions population provided that such pressure between the relative selectivity of preda- was largely additive; however, models tion and the degree to which mortality in predicted that sufficiently strong selection infected deer is compensatory also will could still dampen the dynamics of influence epidemic dynamics in emergent emergent CWD epidemics even in cases CWD foci (Fig. 3). Predicted CWD prev- where mortality among infected deer was alence 20 yr after introducing a single largely compensatory. infected deer into simulated populations subjected to 15% annual predation under DISCUSSION different combinations of selectivity and compensation varied from ,0to8%.In Results from these simulations suggest general, simulations suggested that even that predation could markedly decrease modest levels of selectivity might be prevalence of CWD under certain condi- 84 JOURNAL OF WILDLIFE DISEASES, VOL. 47, NO. 1, JANUARY 2011 FIGURE 2. Results of model experiments to examine sensitivity of variation in uncertain model parameters. Open circles show results from simulations with no predation, diamonds show nonselective predation at a rate of 0.15. A. Effect of variation in prey vulnerability to selective predation (v52, solid line; 4, dashed line; 8, dotted line). Increasing values of v indicate greater selection for infected over susceptible animals. In all cases, compensation was held constant at 0.3 and predation rate at 0.15. B. Effect of variation in the level of compensation between predation and CWD mortality (c50.1, solid line; 0.3, dashed line; 0.6, dotted line). Increasing values of c indicate greater compensation between predation and CWD. In all cases, selectivity was held constant at four and the predation rate at 0.15. C. Effect of variation in predation rate (d; solid line, 0.10; dashed line, 0.30; dotted line, 0.50). In all cases, vulnerability to selective predation was held constant at 4 and compensation at 0.3. D. Effect of variation in the scaling coefficient for the transmission rate (solid line, 20; dashed line, 30; dotted line, 40). The scaling coefficient reduces the transmission rate to account for differences between captive and free-ranging deer. A scaling coefficient of 20 indicates that transmission is 20 time more rapid in captivity than in the wild. For other parameter values, see Table 1 in Appendix A. tions. Nonselective predation, as might more precipitous drop in CWD preva- occur with hunting or culling by humans, lence that would culminate in disease may decrease disease prevalence over elimination in a closed system (Fig. 1E, time but the disease was not eliminated F). Selective predation does not allow a under modeled conditions (Fig. 1C, D). larger population of susceptible animals to Alternatively, selective predation by persist relative to the nonselective case wolves at the same rate would result in a because wolves are assumed to consume WILD ET AL.—PREDATION IN DISEASE CONTROL 85 repeatedly introduced at low levels (i.e., an open population); however, the slope of decline would be variable and elimination might never be achieved because high rates of disease reintroduction may offset selective predation of CWD-positive indi- viduals. Although they are not the most likely scenarios, other combinations of parameters, in particular high excretion rates leading to increased levels of trans- mission, also may result in an inability to eliminate the disease within a reasonable period of time. Simulation results suggested that selec- tive predation could also dampen or FIGURE 3. Predicted CWD prevalence 20 yr eliminate the emergence of CWD in new after introducing a single infected deer into simulat- locations (Fig. 3), adding support to spec- ed populations subjected to 15% annual predation. ulation that the absence of large predators Results of this model experiment revealed interac- presents an amplification risk factor for tions between selectivity and compensation: for establishment of CWD (Samuel et al., example, low compensation and high selective predation result in inability of disease to emerge. 2003). Our prediction may prove testable The benefits of selectivity for reducing prevalence in the future as geographic distribution of are opposed by increasing compensation. CWD expands to areas such as the Greater Yellowstone Ecosystem and more susceptible animals as infected ones northern Wisconsin, USA, and Prince become rare. Although the time required Albert National Park, Canada, where to achieve results depends in a fundamen- wolves are present but adjacent areas lack tal way on assumptions about prey vulner- wolves. The simulated influences of large ability to selective predation and the predators on the outcomes of CWD nature of compensation among different epizootics also may lend insight into sources of mortality, as well as parameters circumstances surrounding the original regulating disease transmission, it appears emergence of CWD in Colorado, where that prevalence could be halved within a wolves have been absent since 1943 and decade and eliminated within the century where mountain lion populations were through sustained predation by a pack of suppressed by bounty hunting at the time wolves that removed 15% of deer per year of likely CWD emergence in the mid- in a closed population. 1900s (Barrows and Holmes, 1990; Miller Although uncertainty in parameter esti- et al., 2000). The origins of CWD are mates limits our confidence in predicting unknown but may have been a result of the precise timeframe required for control spillover of scrapie from domestic sheep or elimination of disease, these time or may represent a spontaneous, naturally estimates provide a basis for comparison occurring spongiform encephalopathy of of approaches. What is most clear is a cervids (Williams and Young, 1992; Spra- consistent and robust trend toward de- ker et al., 1997). Regardless, our simula- creasing CWD prevalence in populations tions suggest that had selective predation subject to predation, particularly selective by wolves been present during that period, predation, over a range of parameter CWD may never have been established or estimates (Fig. 2). A similar decreasing detected. In combination with influences trend would be predicted in a population of human-assisted movement of infected subject to predation where CWD was cervids (Williams et al., 2002) and land use 86 JOURNAL OF WILDLIFE DISEASES, VOL. 47, NO. 1, JANUARY 2011 alterations (Farnsworth et al., 2005), the mule deer compared to hunter-harvested absence of large predators, particularly deer (Krumm et al., 2005). It follows that wolves, over much of their native range in infected deer also would be less attentive the United States (Laliberte and Ripple, to predators, and in later stages, that 2004) has likely played a significant role in emaciation and weakness would decrease the current unnatural distribution and both their fight and flight response prevalence of this disease. capabilities (Krumm et al., 2005, 2009; The decrease in CWD prevalence Miller et al., 2008); a nearly fourfold observed in simulations with selective greater relative risk of infected mule deer predation is most likely a result of succumbing to mountain lion predation removing infectious individuals earlier in (Miller et al., 2008) supports this notion. the disease course. Chronic wasting dis- Furthermore, predators—particularly ease exhibits a prolonged disease course of coursing predators such as wolves—focus about 18–36 mo (Williams and Miller, on animals vulnerable due to odd behavior 2002). Transmission models (Miller et al., or compromised body condition (Temple, 2006) reveal little support for a disease 1987; Mech et al., 1991). Field observa- latency period and instead support early tions also suggest that predators can select onset of prion shedding, potentially from CWD-infected deer: mule deer killed by peripheral lymphoid tissue. Accumulation mountain lions were much more likely of abnormal prion protein (PrPcwd) in deer (odds ratios $3.2) to be infected with has been observed in alimentary tract– CWD than same-sex deer killed in the associated lymphoid tissues as early as vicinity by hunters (Krumm et al., 2009). 42 days following experimental oral inoc- Based on the prolonged course of CWD, ulation (Sigurdson et al., 1999) and in the ability of wolves to detect vulnerable tonsils as much as 20 mo prior to death prey, and field observations of mountain from naturally occurring CWD (Wild et lion predation patterns in a system where al., 2002). Moreover, orally inoculated CWD occurs naturally, we believe that deer shed infectious prions in saliva and selective predation modeled at a rate four feces 6–11 mo or more before the onset of times higher than that of healthy deer is a clinical signs (Mathiason et al., 2009; reasonable, if not conservative, estimate. Tamgu¨ney et al., 2009). Therefore, early Overall, our modeling results also are removal of infected individuals should likely a conservative portrayal of the markedly truncate CWD shedding and beneficial impacts that selective predation resultant opportunities for disease trans- could have on damping prion epidemic mission. dynamics in deer. The model we devel- The prolonged clinical course and type oped did not include carcasses of infected of clinical abnormalities associated with deer as a source of infectivity because CWD make it the prototypic disease for necessary parameter estimates were not selection by predators. Chronic wasting available (Miller et al., 2006). However, disease produces subtle changes in behav- carcasses of CWD-infected deer would be ior and body condition that progress over an added source of environmental infec- weeks or months to overt signs of end- tivity in natural systems (Miller et al., stage disease typified by loss of attentive- 2004), and thus their consumption by ness or response to external stimuli, wolves or other carnivores either via emaciation, and weakness (Williams and selective predation or scavenging would Young, 1980, 1992; Wild et al., 2002). Loss be expected to reduce the contribution of of attentiveness and cognitive function carcass material to the overall pool of due to the neurodegenerative process environmental infectivity through local likely account for the marked increase in dispersal and dilution (Krumm et al., risk for vehicle collision of CWD infected 2009). Passage through the alimentary WILD ET AL.—PREDATION IN DISEASE CONTROL 87 tract of wolves likely markedly degrades from consumption of CWD-positive rumi- infectivity of tissues. In sheep, in vitro nants is unknown; however, no evidence incubation of a dilute scrapie brain of naturally occurring CWD has been inoculum with alimentary tract fluids reported outside four species in the family resulted in almost complete degradation Cervidae. Limited surveillance of preda- of PrP (Jeffrey et al., 2006). Moreover, tors and scavengers in CWD-affected changes in deer behavior due to the areas (Jennelle et al., 2009; Miller and presence of predators, i.e., predation risk Wild, unpubl. data) has not revealed effects or what has been termed the evidence of abnormal prion accumulation. ecology of fear (Brown et al., 1999; Ripple Naturally occurring transmissible spongi- and Beschta, 2004), include changes in form encephalopathies (TSE) other than use of space through habitat preferences CWD have been documented in domestic or foraging patterns within a given habitat, mink (transmissible mink encephalopa- or both (Lima and Dill, 1990). If deer thy), domestic sheep and goats (scrapie), move more within established home and domestic cattle (bovine spongiform ranges due to fear of predation, then encephalopathy [BSE]), as well as in contact rates with environmental deposits humans (variant Creutzfeldt-Jacob dis- of infectivity also might diminish. Given ease) and domestic and captive wild felids the sensitivity of epidemic dynamics to (feline spongiform encephalopathy) that such contact rates, even relatively small consumed BSE-contaminated feed (Ho¨rn- reductions would further dampen epidem- limann et al., 2007). Interestingly howev- ic dynamics beyond effects arising from er, no TSE has been observed in a canid selective predation on infected deer alone. despite dietary challenge of BSE to dogs Although here we modeled wolf preda- (Kirkwood and Cunningham, 1994). A tion on deer, similar outcomes would be species barrier is generally believed to be expected for wolf predation on other responsible for the specificity of prion species susceptible to CWD. Hobbs diseases to their respective hosts, although (2006) used CWD and elk (Cervus ela- some spillover, as with BSE, has been phus nelsoni) population data from Rocky documented for at least one prion strain. Mountain National Park (Colorado, USA) Raymond et al. (2000) demonstrated a to model the impact on CWD that may be barrier at the molecular level that they achieved through maintaining a pack of suggest limits the susceptibility of non- wolves in the park. Results from these cervid species to CWD. The dog and wolf simulations supported the idea that pre- are very similar in PrP sequence and quite dation could drive decreases in CWD different from cattle, domestic cats, and prevalence over a range of parameter elk (Scha¨tzl, 2007). estimates. Impacts by predators other than We suggest that predation, particularly wolves may also reduce CWD prevalence wolf predation, may be a useful tool for to varying degrees, as seen in our results management of CWD. Currently, the from nonselective removal by humans. We range of wolves (Boitani, 2003) does not consider the wolf, a large coursing pred- overlap with the distribution of CWD ator, to be most effective in selective (Chronic Wasting Disease Alliance, 2009) removal of deer vulnerable from CWD so our predictions on the effects of wolves infection; however, opportunistic moun- on CWD prevalence remain untested. tain lions (Krumm et al., 2009), and However, as wolf range expands through potentially coyote (Canis latrans) packs, Wyoming and Wisconsin, USA, and Al- would likely benefit from lack of vigilance berta and Saskatchewan, Canada, and into by CWD-affected deer as well. Colorado and Utah, USA, the possibility The potential impact on wolves and for such evaluation may occur. Alterna- other native North American predators tively, CWD may be detected in a new 88 JOURNAL OF WILDLIFE DISEASES, VOL. 47, NO. 1, JANUARY 2011 geographic location where wolves are funding from the US Department of Agricul- present. Based on our simulations, disease ture, Animal and Plant Health Inspection Service, Veterinary Services. We thank A. may be difficult to detect in these areas Graham, D. Reep, D. Kammerer, and J. Spaak unless unique methods of surveillance, for administrative support. M. Rice and B. such as monitoring of wolf-killed cervids Walker provided helpful comments on an for presence of PrPres, are implemented. earlier manuscript draft. Beschta and Ripple (2009) suggest that LITERATURE CITED restoration of large predators, such as wolves, provides a recovery strategy for ALTIZER, S., C. L. NUNN,P.H.THRALL,J.L. native flora, functional predator-prey- GITTLEMAN,J.ANTONOVICS,A.A.CUNNINGHAM, A. P. DOBSON,V.EZENWA,K.E.JONES,A.B. scavenger food webs, and ecosystems PEDERSEN,M.POSS, AND J. R. C. PULLIAM. 2003. degraded by overabundant wild ungulates. Social organization and parasite risk in mam- Wolf restoration also provides an oppor- mals: Integrating theory and empirical studies. tunity to observe and evaluate the effects Annual Review of Ecology, Evolution, and that selective predation may have on Systematics 34: 517–547. BAETEN, L. A., B. E. POWERS,J.E.JEWELL,T.R. prevalence of an invariably fatal chronic SPRAKER, AND M. W. MILLER. 2007. A natural disease in deer and elk. In areas where case of chronic wasting disease in a free-ranging predator restoration is not possible, de- moose (Alces alces shirasi). Journal of Wildlife ployment of wolves as stewardship tools Diseases 43: 309–314. for the primary purpose of disease control BARLOW, N. D. 1996. The ecology of wildlife disease control: Simple models revisited. Journal of could provide a novel approach to man- Applied Ecology 33: 303–314. agement. BARROWS, P., AND J. HOLMES. 1990. Colorado’s Although somewhat novel, the concept wildlife story. Colorado Division of Wildlife, of using wildlife species as stewardship Denver, Colorado. pp. 450. tools to provide ecosystem services is not BESCHTA, R. L., AND W. J. RIPPLE. 2009. Large predators and trophic cascades in terrestrial new. Restoration of bison to reestablish ecosystems of the western United States. Bio- healthy landscapes of prairie vegetation in logical Conservation 142: 2401–2414. doi:10.1016/ the United States (United States Depart- j.biocon.2009.06.015. ment of the Interior, 2008) and large BOITANI, L. 2003. Wolf conservation and recovery. In carnivore, (e.g., lion [Panthera leo]), trans- Wolves: Behavior, ecology, and conservation, L. D. Mech and L. Boitani (eds.). University of locations to restore ecologic integrity in Chicago Press, Chicago, Illinois. pp. 317–340. fenced parks in Africa (Hayward et al., BROWN, J. S., J. W. LAUNDRE, AND M. GURUNG. 1999. 2007) are occurring. Licht et al. (2010) The ecology of fear: Optimal foraging, game propose use of small populations of wolves theory, and trophic interactions. Journal of Mammalogy 80: 385–399. for ecosystem restoration in North Amer- CHOISY, M., AND P. ROHANI. 2006. Harvesting can in- ica. Public tolerance of wildlife, particu- crease severity of wildlife disease epidemics. Pro- larly predators, may dictate intensive ceedings of the Royal Society B: 1–10. doi:10.1098/ management in species used in such rspb.2006.3554. restoration efforts. Regardless of whether CHOO, K., P. D. WILLIAMS, AND T. DAY. 2003. Host mortality, predation, and the evolution of wolves are managed under natural regu- parasite virulence. Ecology Letters 6: 310–315. lation or primarily for fulfilling their CONNER, M. M., M. W. MILLER,M.R.EBINGER, AND ecologic role, they provide a promising K. P. BURNHAM. 2007. A meta-BACI approach approach for control of CWD that war- for evaluating management intervention on rants further evaluation. chronic wasting disease in mule deer. Ecological Applications 17: 140–153. ACKNOWLEDGMENTS CHRONIC WASTING DISEASE ALLIANCE. 2009. Welcome to the Chronic Wasting Disease Alliance’s web Our work was supported by the US National site. www.cwd-info.org. Accessed November Park Service and National Science Founda- 2009. tion/National Institutes of Health grants DEB- DASZAK, P., A. A. CUNNINGHAM, AND A. D. HYATT. 0091961 and EF-0914489, with supplemental 2000. Emerging infectious diseases of wildlife— WILD ET AL.—PREDATION IN DISEASE CONTROL 89 Threats to biodiversity and human health. Do parasites make prey vulnerable to predation? Science 287: 443–449. Red grouse and parasites. Journal of Animal ———, ———, AND ———. 2001. Anthropogenic Ecology 61: 681–692. environmental change and the emergence of HO¨ RNLIMANN, B., D. RIESNER,H.KRETZSCHMAR,R.G. infectious diseases in wildlife. Acta Tropica 78: WILL,S.C.MACDIARMID,G.A.H.WELLS, AND 103–116. M. P. ALPERS. 2007. History. In Prions in animals FARNSWORTH, M. L., L. L. WOLFE,N.T.HOBBS,K.P. and humans, B. Ho¨rnlimann, D. Riesner and H. BURNHAM,E.S.WILLIAMS,D.M.THEOBALD, Kretzschmar (eds.). Walter de Gruyter GmbH & M. M. CONNER, AND M. W. MILLER. 2005. Co. KG, Berlin, Germany. pp. 3–27. Human land use influences chronic wasting JEFFREY, M., L. GONZALEZ,A.ESPENES,C.M.PRESS, disease prevalence in mule deer. Ecological S. MARTIN,M.CHAPLIN,L.DAVIS,T.LNADSVERK, Applications 15: 119–126. C. MACALDOWIE,S.EATON, AND G. MCGOVERN. FOX, K. A., J. E. JEWELL,E.S.WILLIAMS, AND M. W. 2006. Transportation of prion protein across the CWD MILLER. 2006. Patterns of PrP accumulation intestinal mucosa of scrapie-susceptible and during the course of chronic wasting disease scrapie-resistant sheep. Journal of Pathology infection in orally inoculated mule deer (Odo- 209: 4–14. coileus hemionus). Journal of General Virology JENNELLE, C. S., M. D. SAMUEL,C.A.NOLDEN,D.P. 87: 3451–3461. KEANE,D.J.BARR,C.JOHNSON,J.P.VANDERLOO, GREAR, D. A., M. D. SAMUEL,J.A.LANGENBERG, AND J. M. AIKEN,A.N.HAMIR, AND E. A. HOOVER. D. KEANE. 2006. Demographic patterns and 2009. Surveillance for transmissible spongiform harvest vulnerability of chronic wasting disease encephalopathy in scavengers of white-tailed infected white-tailed deer in Wisconsin. Journal deer carcasses in the chronic wasting disease of Wildlife Management 70: 546–553. area of Wisconsin. Journal of Toxicology and GROSS, J. E., AND M. W. MILLER. 2001. Chronic Environmental Health, Part A 72: 1018–1024. wasting disease in mule deer: Disease dynamics JOHNSON, P. T. J., J. M. CHASE,K.L.DOSCH,R.B. and control. Journal of Wildlife Management 65: HARTSON,J.A.GROSS,D.J.LARSON,D.R. 205–215. SUTHERLAND, AND S. R. CARPENTER. 2007. Aquat- HARVELL, C. D., K. KIM,J.M.BURKHOLDER, J. M., R. ic eutrophication promotes pathogenic infection R. COLWELL,P.R.EPSTEIN,D.J.GRIMES,E.E. in amphibians. Proceedings of the National HOFMANN,E.K.LIPP,A.D.M.E.OSTERHAUS, Academy of Sciences of the United States of R. M. OVERSTREET,J.W.PORTER,G.W.SMITH, America 104: 15781–15786. AND G. R. VASTA. 1999. Emerging marine JOLY, D. O., AND F. MESSIER. 2004a. The distribution diseases—Climate links and anthropogenic fac- of Echinococcus granulosus in moose: Evidence tors. Science 285: 1505–1510. for parasite-induced vulnerability to predation HAYWARD, M. W., J. ADENDORFF, J. O’BRIEN,A. by wolves? Oecologia 140: 586–590. SHOLTO-DOUGLAS,C.BISSETT,L.C.MOOLMAN,P. ———, AND ———. 2004b. Testing hypotheses of BEAN,A.FOGARTY,D.HOWARTH,R.SLATER, AND bison population decline (1970–1999) in Wood G. I. H. KERLEY. 2007. Practical considerations Buffalo National Park: Synergism between exotic for the reintroduction of large, terrestrial, disease and predation. Canadian Journal of mammalian predators based on reintroductions Zoology 82: 1165–1176. to South Africa’s Eastern Cape Province. Open KIRKWOOD, J. K., AND A. A. CUNNINGHAM. 1994. Conservation Biology Journal 1: 1–11. Epidemiological observations on spongiform HEESTERBEEK, J. A. P., AND M. G. ROBERTS. 1995. encephalopathies in captive wild animals in the Mathematical models for microparasites of British Isles. Veterinary Record 135: 296–303. wildlife. In Ecology of infectious diseases in KRUMM, C. E., M. M. CONNER, AND M. W. MILLER. natural populations, B. T. Grenfell and A. P. 2005. Relative vulnerability of chronic wasting Dobson (eds.). Cambridge University Press, disease infected mule deer to vehicle collision. Cambridge, UK. pp. 90–122. Journal of Wildlife Diseases 41: 503–511. HOBBS, N. T. 1996. Modification of ecosystems by ———, ———, N. T. HOBBS,D.O.HUNTER, AND M. ungulates. Journal of Wildlife Management 60: W. MILLER. 2009. Mountain lions prey selec- 695–713. tively on prion-infected mule deer. Biology ———. 2006. A model analysis of effects of wolf Letters. doi: 10.1098/rsbl.2009.0742. predation on prevalence of chronic wasting KUTZ, S. J., E. P. HOBERG,L.POLLEY, AND E. J. disease in elk populations of Rocky Mountain JENKINS. 2005. Global warming is changing the National Park. National Park Service Report, dynamics of arctic host-parasite systems. Pro- 9 pp. ceedings of the Royal Society B: Biological HOLT, R. D., AND M. ROY. 2007. Predation can Sciences 272: 2571–2576. increase the prevalence of infectious disease. LAFFERTY, K. D., AND R. D. HOLT. 2003. How should American Naturalist 169: 690–699. environmental stress affect the population dy- HUDSON, P. J., A. P. DOBSON, AND D. NEWBORN. 1992. namics of disease? Ecology Letters 6: 654–664. 90 JOURNAL OF WILDLIFE DISEASES, VOL. 47, NO. 1, JANUARY 2011 ———, AND A. K. MORRIS. 1996. Altered behavior of ———, ———, C. W. MCCARTY,T.R.SPRAKER,T.J. parasitized killifish increases susceptibility to KREEGER,C.T.LARSEN, AND E. T. THORNE. 2000. predation by bird final hosts. Ecology 77: Epizootiology of chronic wasting disease in free- 1390–1397. ranging cervids in Colorado and Wyoming. LAIOLO, P., D. SERRANO,J.L.TELLA,M.CARRETE,G. Journal of Wildlife Diseases 36: 676–690. LOPEZ, AND C. NAVARRO. 2007. Distress calls ———, ———, N. T. HOBBS, AND L. L. WOLFE. reflect poxvirus infection in lesser short-toed lark 2004. Environmental sources of prion transmis- Calandrella rufescens. Behavioral Ecology 18: sion in mule deer. Emerging Infectious Diseases 507–512. 10: 1003–1006. LALIBERTE, A. S., AND W. J. RIPPLE. 2004. Range ———, N. T. HOBBS, AND S. J. TAVENER. 2006. contractions of North American carnivores and Dynamics of prion disease transmission in mule ungulates. BioScience 54: 123–138. deer. Ecological Applications 16: 2208–2214. LEFCORT, H., AND A. R. BLAUSTEIN. 1995. Disease, ———, H. M. SWANSON,L.L.WOLFE,F.G. predator avoidance, and vulnerability to preda- QUARTARONE,S.L.HUWER,C.H.SOUTHWICK, tion in tadpoles. Oikos 74: 469–474. AND P. M. LUKACS. 2008. Lions and prions and LEVINS, R. 1966. The strategy of model building in deer demise. PLoS ONE 3(12): e4019. doi:10.1371/ population biology. American Scientist 54: 421– journal.pone.0004019. 431. MOLLER, A. P., AND J. ERRITZOE. 2000. Predation LICHT, D. S., J. J. MILLSPAUGH,K.E.KUNKEL,C.O. against birds with low immunocompetence. KOCHANNY, AND R. O. PETERSON. 2010. Using Oecologia 122: 500–504. small populations of wolves for ecosystem ———, AND J. T. NIELSEN. 2007. Malaria and risk of restoration and stewardship. BioScience 60: predation: A comparative study of birds. Ecology 147–153. 88: 871–881. LIMA, S. L., AND L. M. DILL. 1990. Behavior decisions MURRAY, D. L., J. R. CARY, AND L. B. KEITH. 1997. made under the risk of predation: A review and Interactive effects of sublethal nematodes and prospectus. Canadian Journal of Zoology 68: nutritional status on snowshoe hare vulnerability 619–640. to predation. Journal of Animal Ecology 66: 250– MATHIASON, C. K., J. G. POWERS,S.J.DAHMES,D.A. 264. OSBORN,K.V.MILLER,R.J.WARREN,G.L. OSTFELD, R. S., AND R. D. HOLT. 2004. Are predators MASON,S.A.HAYS,J.HAYES-KLUG,D.M. good for your health? Evaluating evidence for SEELIG,M.A.WILD,L.L.WOLFE,T.R.SPRAKER, top-down regulation of zoonotic disease reser- M. W. MILLER,C.J.SIGURDSON,G.C.TELLING, voirs. Frontiers in Ecology and the Environment AND E. A. HOOVER. 2006. Infectious prions in the 2: 13–20. saliva and blood of deer with chronic wasting PACKER, C., R. D. HOLT,P.J.HUDSON,K.D. disease. Science 314: 133–136. LAFFERTY, AND A. P. DOBSON. 2003. Keeping the ———, S. A. HAYS,J.POWERS,J.HAYES-KLUG,J. herds healthy and alert: Implications of predator LANGENBERG,S.J.DAHMES,D.A.OSBORN,K.V. control for infectious disease. Ecology Letters 6: MILLER,R.J.WARREN,G.L.MASON, AND E. A. 797–802. HOOVER. 2009. Infectious prions in pre-clinical PEDERSEN, A. B., K. E. JONES,C.L.NUNN, AND S. deer and transmission of chronic wasting disease ALTIZER. 2007. Infectious diseases and extinction solely by environmental exposure. PLoS ONE risk in wild mammals. Conservation Biology 21: 4(6): e5916. doi:10.1371/journal.pone.0005916. 1269–1279. MECH, L. D., AND R. O. PETERSON. 2003. Wolf-prey RDEVELOPMENT CORE TEAM. 2008. R: A language relations. In Wolves: Behavior, ecology, and and environment for statistical computing. R conservation, L. D. Mech and L. Boitani (eds.). Foundation for Statistical Computing, Vienna, University of Chicago Press, Chicago, Illinois. Austria, http://www.R-project.org. pp. 131–160. RAYMOND, G. J., A. BOSSERS,L.D.RAYMOND,K.I. ———, M. E. NELSON, AND R. E. MCROBERTS. 1991. O’ROURKE,L.E.MCHOLLAND,P.K.BRYANT, III, Effects of maternal and grandmaternal nutrition M. W. MILLER,E.S.WILLIAMS,M.SMITS, AND B. on deer mass and vulnerability to wolf predation. CAUGHEY. 2000. Evidence of a molecular barrier Journal of Mammalogy 72: 146–151. limiting susceptibility of humans, cattle and MEDIN, D. E., AND A. E. ANDERSON. 1979. Modeling sheep to chronic wasting disease. EMBO Journal the dynamics of a Colorado mule deer popula- 19: 4425–4430. tion. Wildlife Monographs 68: 3–77. RIPPLE, W. J., AND R. L. BESCHTA. 2004. Wolves and MILLER, M. W., AND M. A. WILD. 2004. Epidemiology the ecology of fear: Can predation risk structure of chronic wasting disease in captive white-tailed ecosystems? BioScience 54: 755–766. and mule deer. Journal of Wildlife Diseases 40: SAMUEL, M. D., D. O. JOLY,M.A.WILD,S.D. 320–327. WRIGHT,D.L.OTIS,R.W.WERGE, AND M. W. ———, AND E. S. WILLIAMS. 2003. Horizontal prion MILLER. 2003. Surveillance strategies for detect- transmission in mule deer. Nature 425: 35–36. ing chronic wasting disease in free-ranging deer WILD ET AL.—PREDATION IN DISEASE CONTROL 91 and elk. US Geological Survey, National Wildlife 1998. Heteroxenous coccidian increase the Health Center, Madison, Wisconsin. 41 pp. predation risk of parasitized rodents. Parasitol- SCHA¨ TZL, H. M. 2007. The phylogeny of mammalian ogy 117: 521–524. and nonmammalian prion proteins. In Prions in WHITE, G. C., AND R. M. BARTMANN. 1998. Effect of animals and humans, B. Ho¨rnlimann, D. Riesner, density reduction on overwinter survival of free- and H. Kretzschmar (eds.). Walter de Gruyter ranging mule deer fawns. Journal of Wildlife GmbH & Co. KG, Berlin, Germany. pp. 119– Management 62: 214–225. 133. WILD, M. A., T. R. SPRAKER,C.J.SIGURDSON,K.I. SIGURDSON, C. J., E. S. WILLIAMS,M.W.MILLER,T. O’ROURKE, AND M. W. MILLER. 2002. Preclinical R. SPRAKER,K.I.O’ROURKE, AND E. A. HOOVER. diagnosis of chronic wasting disease in captive 1999. Oral transmission and early lymphoid mule deer (Odocoileus hemionus) and white- tropism of chronic wasting disease PrPres in tailed deer (Odocoileus virginianus) using ton- mule deer fawns (Odocoileus hemionus). Journal sillar biopsy. Journal of General Virology 83: of General Virology 80: 2757–2764. 2629–2634. SPRAKER, T. R., M. W. MILLER,E.S.WILLIAMS,D.M. WILLIAMS, E. S., AND M. W. MILLER. 2002. Chronic GETZY,W.J.ADRIAN,G.G.SCHOONVELD,R.A. wasting disease in deer and elk in North SPOWART,K.I.O’ROURKE,J.M.MILLER, AND America. In Infectious diseases of wildlife: P. A. MERZ. 1997. Spongiform encephalopathy Detection, diagnosis, and management, R. G. in free-ranging mule deer (Odocoileus hemi- Bengis (ed.). Revue scientifique et technique onus), white-tailed deer (Odocoileus virginianus) Office international des Epizooties 21: 305–316. and Rocky Mountain elk (Cervus elaphus ———, AND S. YOUNG. 1980. Chronic wasting disease nelsoni) in northcentral Colorado. Journal of of captive mule deer: A spongiform encephalop- Wildlife Diseases 33: 1–6. athy. Journal of Wildlife Diseases 16: 89–98. TAMGU¨ NEY, G., M. W. MILLER,L.L.WOLFE,T.M. ———, AND ———. 1992. Spongiform encephalop- SIROCHMAN,D.V.GLIDDEN,C.PALMER,A. athies of Cervidae. In Transmissible spongiform LEMUS,S.J.DEARMOND, AND S. B. PRUSINER. encephalopathies of animals, R. Bradley and D. 2009. Asymptomatic deer excrete infectious Mathews (eds.). Revue scientifique et technique prions in faeces. Nature 461: 529–532, Office international des Epizooties 11: 551–567. doi:10.1038/nature08289. ———, M. W. MILLER,T.J.KREEGER,R.H.KAHN, TEMPLE, S. A. 1987. Do predators always capture AND E. T. THORNE. 2002. Chronic wasting substandard individuals disproportionately from disease of deer and elk: A review with recom- prey populations? Ecology 68: 669–674. mendations for management. Journal of Wildlife UNITED STATES DEPARTMENT OF THE INTERIOR. 2008. Management 66: 551–563. Bison Conservation Initiative. US Department WOLFE, L. L., M. W. MILLER, AND E. S. WILLIAMS. of the Interior, Washington, D.C., 11 pp. www. 2004. Feasibility of ‘‘test–and–cull’’ for managing interior.gov/initiatives/bison/Bison%20Bridge%20 chronic wasting disease in urban mule deer. Page%20DOI%20 Bison%20Conservation%20 Wildlife Society Bulletin 32: 500–505. Initiative%20framework.pdf. Accessed 20 November 2009. Submitted for publication 4 December 2009. VORˇ ı´Sˇ EK, P., J. VOTYPKA,K.ZVARA, AND M. SVOBODOVA. Accepted 7 October 2010. APPENDIX A—MODEL STRUCTURE dS ~aIðÞzS {S(cEzm), Using data from two epidemics of chronic dt wasting disease (CWD) in a captive population dI ~cSE{I(mzm), ð1Þ of mule deer (Odocoileus hemionus), Miller et dt al. (2006) found that models of indirect dE transmission of CWD from excreta had almost ~eI{tE, seven times more support in data than more dt traditional models of direct, animal-to-animal transmission. The best approximating model in where their studies used three linked differential S5number of susceptible (uninfected) ani- equations representing the number of infected mals, I5number of infected animals, E5the and susceptible animals and the mass of mass of infectious material in the environ- infectious material in the environment: ment, a5the per capita birth rate, m5the per 92 JOURNAL OF WILDLIFE DISEASES, VOL. 47, NO. 1, JANUARY 2011 capita death rate from causes other than selectively kill twice as many infected animals CWD, c5the indirect transmission coefficient, as susceptible ones. A value of v51 indicates m5the additive, per capita death rate from no vulnerability of infected animals and CWD, e5the per capita rate of excretion of increasing values of v above 1 indicate infectious material by infected animals, and increasing vulnerability to selective predation. t5the mass specific rate of loss of infectious If predators select prey totally at random, material from the environment. then the probability of dying from CWD is This model is based on two assumptions, independent of the probability of dying from that the instantaneous per capita rate of predation, as d is defined. In this case the infection was directly proportionate to the probability that an infected animal will survive, mass of infectious material in the environment Q, over an interval of time5Dt is (i.e., dI/dtS5cE) and that the rate of uptake of {ðÞz z infectious material by deer has negligible w~e m m d Dt: ð4Þ effects on the pool size. We modified this model to include density- However, when predators are selective, then it dependent effects on recruitment into the follows by definition that the probability of population and to include selective and dying from predation is not independent of the nonselective predation: probability of dying from the disease: dS SzI ~ {½ mzmzd(1{c) Dt ð Þ ~aSðÞzI 1{ {S(cEzm) w e 5 dt Ka { { ðÞz The term c allows us to represent the extent to (1 p)d S I , which predation mortality compensates for ð2Þ dI CWD mortality. Because 1/m+m+d is the ~cSE{I(mzm){p(1{c)dðÞSzI , average lifetime of an infected animal assum- dt ing that disease mortality and predation dE ~eI{tE, mortality are completely additive, it follows dt that 1/[m+m+d(12c)]21/(m+m+d) is the in- crease in the average lifetime of an infected where Ka is the population level where birth animal that results because predation mortality 5 d rate 0 and is the additive, instantaneous per may not fully add to disease mortality. The capita rate of predation when predators select value of c ranges from 0 to 1. When c50, then prey randomly. Predation rates were adjusted predation mortality is completely additive to account for selectivity by the term p, which with CWD mortality, as in equation (4). When represents the proportion of the total kill that c51, predation mortality is completely com- was infected. We calculated p as pensatory and does not add to deaths from vI disease (i.e., deer would have died from CWD p~ ð3Þ vIzS within the year had they not been preyed upon). where v is the vulnerability of infected animals To solve the system of equations in (2), we relative to susceptible ones. Relative vulnera- used numeric integration implemented in the bility is a multiplier giving the number of lsoda package of the R computing environ- infected animals in the total kill per suscepti- ment (R Development Core Team, 2008). ble animal, assuming equal abundance of Values for parameters used in simulations are infected and susceptible. Thus, a value of derived from Miller et al. (2006) and plausible v52 means that if susceptible and infected assumptions about deer population dynamics animals were equally abundant, wolves would in the absence of CWD (Table 1). WILD ET AL.—PREDATION IN DISEASE CONTROL 93 TABLE 1. Values for model parameters used in example simulations. Parameter Definition Valuea Reference or source a Birth rateb at population50 0.6 Medin and Anderson, 1979 m Non-CWDc death rated 0.1 White and Bartmann, 1998 Ka Population at which birth 1,230 Assigned rate50 c Transmission ratee 0.787 Miller et al., 2006 m CWD death rate 0.567 Miller et al., 2006 e Rate of excretion of infec- 0.111 Miller et al., 2006 tious material t Rate of loss of infectious 2.55 Miller et al., 2006 material from the envi- ronment a Units for all rates are per year. b The birth rate in continuous time, which corresponds to a discrete time birth rate of 1.8 fawns per female. c CWD5chronic wasting disease. d The continuous-time death rate corresponds to an annual adult survival probability of 0.90. e The transmission rate was scaled to account for differences in density between the wild and the captive setting where it was measured by Miller et al. (2006). The default scaling factor was allowed for densities in paddocks that were 25 times higher than in the wild, thus 0.787/25. Open access, freely available online PLoS BIOLOGY Gray Wolves as Climate Change Buffers in Yellowstone Christopher C. Wilmers1*, Wayne M. Getz1,2 1 Department of Environmental Science, Policy and Management, University of California, Berkeley, California, United States of America, 2 Mammal Research Institute, Department of Zoology and Entomology, University of Pretoria, South Africa Understanding the mechanisms by which climate and predation patterns by top predators co-vary to affect community structure accrues added importance as humans exert growing influence over both climate and regional predator assemblages. In Yellowstone National Park, winter conditions and reintroduced gray wolves (Canis lupus) together determine the availability of winter carrion on which numerous scavenger species depend for survival and reproduction. As climate changes in Yellowstone, therefore, scavenger species may experience a dramatic reshuffling of food resources. As such, we analyzed 55 y of weather data from Yellowstone in order to determine trends in winter conditions. We found that winters are getting shorter, as measured by the number of days with snow on the ground, due to decreased snowfall and increased number of days with temperatures above freezing. To investigate synergistic effects of human and climatic alterations of species interactions, we used an empirically derived model to show that in the absence of wolves, early snow thaw leads to a substantial reduction in late-winter carrion, causing potential food bottlenecks for scavengers. In addition, by narrowing the window of time over which carrion is available and thereby creating a resource pulse, climate change likely favors scavengers that can quickly track food sources over great distances. Wolves, however, largely mitigate late-winter reduction in carrion due to earlier snow thaws. By buffering the effects of climate change on carrion availability, wolves allow scavengers to adapt to a changing environment over a longer time scale more commensurate with natural processes. This study illustrates the importance of restoring and maintaining intact food chains in the face of large-scale environmental perturbations such as climate change. Citation: Wilmers CC, Getz WM (2005) Gray wolves as climate change buffers in Yellowstone. PLoS Biol 3(4): e92. Introduction to the perturbing effects of climate change than more speciose communities. As such, understanding the mecha- Average earth temperatures have increased by 0.6 8C over nisms or pathways that confer community resistance to the last 100 years [1] and are predicted to increase by 1.4–5.8 climate change will be important to conservationists and 8C over the next century [2]. Commensurate with rising global managers in mitigating the effects of a changing climate on temperatures are regional changes in weather patterns shifting community patterns and local extinctions. affecting the quantity and timing of precipitation and The reintroduction of gray wolves (Canis lupus) to Yellow- moisture levels. A challenge facing ecologists is to understand stone National Park (NP) in 1995 [15] provides a research how these changes in the abiotic environment will impact opportunity for comparing the response of an ecosystem to populations and communities of organisms. Already, studies climate change in scenarios with and without direct human have documented the effect of a changing climate on the alteration of species composition. Wolf restoration is already phenology, range, reproductive success, and synchrony of realizing a change on the Yellowstone ecosystem by altering certain plants and animals (see [1] for a comprehensive the quantity and timing of carrion availability to scavengers review). In addition, climate-caused community-level changes [16]. Ravens (Corvus corax), bald eagles (Haliaeetus leucocephalus), have been documented when range shifts lead to the transfer golden eagles (Aquila chrysaetos), magpies (Pica pica), coyotes of an entire assemblage of species [3]. (Canis latrans), grizzly bears (Ursus arctos), and black bears Given such responses by individual species, we can expect (Ursus americanus) are each frequent visitors at wolf kills [17] consequent shifts in trophic structure and competitive and are highly reliant on winter carrion for survival and hierarchies at the community scale [4]. Studies addressing reproductive success [16,18,19,20,21,22]. this problem have focused primarily on how species-specific Prior to wolf reintroduction, winter mortality of elk (Cervus responses in phenology and geographic range alter compet- itive balances and the timing of food availability for neonates [5,6,7,8]. In Britain, for instance, winter warming has Received July 11, 2004; Accepted January 13, 2004; Published March 15, 2005 DOI: 10.1371/journal.pbio.0030092 precipitated disparate responses in the breeding phenology of different amphibian species, exposing frog larvae (Rana Copyright: Ó 2005 Wilmers and Getz. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits temporaria), which have shown no phenological response, to unrestricted use, distribution, and reproduction in any medium, provided the higher levels of predation from newts (Triturus spp.) that are original work is properly cited. entering ponds earlier than before [5]. Abbreviations: NP, National Park; SDTH, snow depth; SNFL, snowfall; TMAX, As predicted by community stability theory, the impact of maximum temperature; TMIN, minimum temperature climate change on communities may vary in relation to levels Academic Editor: Andy P. Dobson, Princeton University, United States of America of species diversity [9,10,11,12]. Depauperate communities or *To whom correspondence should be addressed. E-mail: cwilmers@nature. those lacking keystone species [13,14] may be more vulnerable berkeley.edu PLoS Biology | www.plosbiology.org 0571 April 2005 | Volume 3 | Issue 4 | e92 Climate Change Buffers Figure 1. Winter Snow Depths 1948–2003 at Mammoth Hot Springs Figure 2. Winter Snow Depths 1948–2003 at Tower Falls Average monthly SDTH for November (A), December (B), January (C), Average monthly SDTH for November (A), December (B), January (C), February (D), March (E), and April (F) 1948–2003 at the Mammoth February (D), March (E), and April (F) 1948–2003 at the Tower Falls Hot Springs weather site. weather site. DOI: 10.1371/journal.pbio.0030092.g001 DOI: 10.1371/journal.pbio.0030092.g002 elaphus), the most abundant ungulate in Yellowstone, was in all winter months except November [the effect is largely dependent on snow depth (SDTH) [23]. Deep snows significant at p 0.05 for February through April and nearly lead to increased metabolic activity [24] and decreased access significant for December and January (Figure 1)]. Further- to food resources, thereby causing elk to weaken and die [25]. more, the slope of the line relating SDTH to year becomes In the absence of wolves, carrion was plentiful both during more negative with each month, indicating a more pro- severe winters and at the end of moderate winters, but more nounced effect of climate change in late winter. The result for scarce in early winter or during mild winters [23]. Reintro- April, however, is confounded by a number of zeros, which duced wolves are now the primary cause of elk mortality created a violation of the normality assumption for the linear throughout the year [26]. Scavengers that once relied on regression. Average monthly SDTH at the Tower Falls winter-killed elk for food now depend on kleptoparasitizing weather site (Figure 2) did not indicate a strong pattern in wolf-killed elk [16]. Hence carrion availability has become the early winter, but showed a significant decline in the late- primarily a function of wolf pack size, with SDTH an winter months of March and April (Figure 2E and 2F). important but secondary factor. Winters in Yellowstone are getting shorter. While we did As global temperatures rise, evidence suggests that north- not detect a difference in the date of the arrival of the first ern latitude and high elevation areas will experience shorter snow, we did detect a declining trend in the date of last snow winters and earlier snow melts [27]. Given the overwhelming on the ground (Figure 3A and 3B). influence of gray wolves on scavenger food webs, community- At both the Tower and Mammoth weather sites, the level responses to climatic changes in the absence of wolves number of days that maximum temperature (TMAX) ex- may differ substantially from those in the presence of ceeded freezing for the period of January through March Yellowstone’s newly restored top carnivore. As such, we increased significantly (Figure 3C and 3D). Furthermore, analyzed over 50 y of weather data from Yellowstone’s midwinter snowfall is decreasing, and late-winter minimum northern range for trends in winter conditions, and temperature (TMIN) and TMAX show signs of increasing in constructed empirically and dynamically grounded scenarios certain months (Table 1). to investigate how changes in SDTH and seasonality differ- entially affect scavengers in the presence and absence of Wolf Effects wolves. Statistical model. The presence of wolves in Yellowstone significantly mitigates the reduction in late-winter carrion expected under climate change (Figure 4). In the scenario Results without wolves, late-winter carrion availability is reduced by Weather Data Analysis 27% in March and by 66% in April. In contrast, the scenario Overthepast55y,averagemonthlySDTHatthe with wolves reveals a reduction in carrion availability of only Mammoth Hot Springs weather site show a steady decline 4% in March and 11% in April. There was not a significant PLoS Biology | www.plosbiology.org0572 April 2005 | Volume 3 | Issue 4 | e92 Climate Change Buffers Figure 4. Reduction in Winter Carrion Available to Scavengers due to Climate Change 1950–2000: Statistical Model Shown are percent reductions (6 standard error) in winter carrion Figure 3. Changes in the Last Day of Snow Cover over the Last 55 Years available to scavengers due to climate change from 1950 to 2000 with at Mammoth Hot Springs and Tower Falls and without wolves in our statistical model. * Significant difference Last day of snow cover is reported as the number of days from between the two scenarios. January 1 of that year until the first day of bare ground. Changes in DOI: 10.1371/journal.pbio.0030092.g004 last day of snow cover over the last 55 y are shown for Mammoth Hot Springs (A) and Tower falls (B). The number of days from January through March that temperatures exceeded freezing at Mammoth (C) Discussion and Tower (D) are increasing with time. DOI: 10.1371/journal.pbio.0030092.g003 The winter period on the northern range of Yellowstone NP is shortening. Both late-winter SDTHs and the overall duration of snow cover have decreased significantly since 1948 (see Figures 1–3). There are several potential causes of difference in the reduction of early- to midwinter carrion reduced snow pack. Average TMIN and TMAX values are (December through February) between the two scenarios. increasing in late winter, while midwinter snowfall appears to Dynamic model. Percent change, z, in late-winter carrion be declining (Table 1). Compounding the effects of declining from 1950 to 2000 was not sensitive to changes in any of the snowfalls on SDTH is an increase in the number of winter parameters in either scenario with or without wolves. 2 days with temperatures above freezing (see Figure 3C and Specifically, r values did not exceed 0.02 for any of the 3D). parameters regressed upon z. Mean monthly percent change in carrion availability from 1950 to 2000 under scenarios with Decreases in late-winter snow pack and in the date of last and without wolves reveals a relative reduction in late-winter snow cover imply that elk will recover sooner from the carrion from 1950 to 2000 and an increase in early-winter detrimental stresses of winter: Smaller snow packs allow elk carrion (Figure 5). Note that this change in carrion easier access to food and decrease energy expenditures availability is much less pronounced in the presence than in required for movement. In addition, herbaceous plant growth the absence of wolves. usually begins within a few days to weeks of last snow cover [28], so elk may increase the quality and quantity of food intake earlier in the year, thus shortening the physiologically stressful winter period. These factors are likely to influence the timing and abundance of carrion as late-winter elk Table 1. Regression Analyses Predicting Mean Monthly SNFL, mortality declines. As we demonstrate here, climate change and Average Late-Winter TMIN and TMAX serves to sharply reduce the amount of late-winter carrion available to Yellowstone’s scavengers (see Figure 4). Accord- Site Dependent Month Intercept Slope r 2 p-value ing to our statistical and dynamic models, however, this Variable reduction is much less pronounced in the presence of wolves. In our statistical model, for instance, we found an 11% Tower Falls SNFL February 84 0.04 0.08 0.055 reduction with wolves versus a 66% reduction without wolves TMIN March 148 0.08 0.08 0.04 in April (see Figure 4). Our dynamic model, which incorpo- TMAX March 77 0.06 0.07 0.06 rates wolf and elk population growth, also reveals a decline in Mammoth Hot SNFL December 106 0.05 0.13 ,0.01 Springs late-winter carrion, especially in the absence of wolves January 121 0.06 0.11 0.02 (Figure 5). In contrast to the statistical model, our dynamic February 71 0.03 0.07 0.056 model predicts an increase in early winter carrion, but less so TMIN March 237 0.13 0.18 ,0.01 with wolves. As the winter period shortens, elk that normally TMAX March 118 0.08 0.11 0.02 would die in March and April will increasingly die in the early winter months, November through February. This will lead to Included are results from regression analyses using year as the independent variable to predict dependent variables SNFL, TMIN, and TMAX for given winter months. We present results for p , 0.10. an increasingly pulsed or seasonal carrion resource. It is DOI: 10.1371/journal.pbio.0030092.t001 important to note that our model has more detailed elk than PLoS Biology | www.plosbiology.org0573 April 2005 | Volume 3 | Issue 4 | e92 Climate Change Buffers species consume the bulk of available scavenge. Our analysis suggests that winter carrion in the absence of wolves will become increasingly pulsed during winter. Consequently, areas without wolves may experience an increase in scav- engers with high recruitment abilities. Actual numerical responses by scavenger species to wolf-provided carrion can now be tested in field studies by comparing areas with wolves to those without wolves in order to determine if changes in scavenger population sizes following wolf reintroduction are consistent with the predicted magnitude of the temporal subsidy due to wolves. As the climate warms, those species will persist that are able to adapt to differences in the environment. Late-winter carrion in Yellowstone will decline with or without wolves, but by buffering this reduction, wolves extend the timescale over which scavenger species can adapt to the changing Figure 5. Change in Carrion Available to Scavengers due to Climate environment. It is important to note that under present-day Change 1950–2000: Dynamic Model climatic conditions, we expect wolves to decrease the long- Shown is the mean monthly change (6 standard error) in carrion term average elk population in Yellowstone [29]. This will available to scavengers due to climate change from 1950 to 2000 with lead to a corresponding decrease in average yearly carrion and without wolves in our dynamic model. levels, which is expected to be small, however, because DOI: 10.1371/journal.pbio.0030092.g005 declines in carrion due to a drop in elk numbers will be partly offset by a higher turnover in the elk population due to wolf dynamics. As suitable data become available, future work wolf predation on old animals [29]. Scenarios both with and can attempt to tease out such factors as the effects of SDTH without wolves therefore provide a meaningful and roughly and territoriality on wolf kill-rate. In both our dynamic and equivalent (see Figure 4 in [29]) amount of carrion to statistical models we find that wolves buffer the effects of scavengers. What we demonstrate here is that scavengers in climate change on carrion abundance and timing. areas without wolves will experience carrion as an increas- This effect will be crucial to scavenger species in the ingly pulsed resource under climate change, whereas in areas Yellowstone area that are highly dependent on winter and with wolves carrion will remain spread out over the winter spring carrion for overwinter survival and reproduction. months. Under scenarios without wolves, these species could face food The primary objective of this study is to understand the bottlenecks in the absence of late-winter carrion. The influence of winter climate and predation on trophic magnitude of this effect will depend on how quickly these dynamics. Our analysis is retrospective, examining what species adapt to a changing environment and how their other would have happened to scavenge availability in scenarios food resources respond to a shortening of the winter period. with and without wolves over the last fifty years of climate Asynchrony of organismal responses to climate change has change. One may ask, however, what these results imply in been prevalent in other areas, leading to changes in the light of predictions for continuing global warming into the competitive balance between species and to food shortages at future. Elk population numbers in Yellowstone are currently important times of year [1]. Yellowstone should prove no constrained by the availability of winter range, where snow exception. Species that respond to weather cues, such as levels are low enough to allow for elk movement and many herbaceous plants, will simply start growing earlier in cratering through the snow to access food resources. If snow the year in response to earlier snow melt. Species that levels in Yellowstone continue to decline in the future, winter respond primarily to day length cues, such as some hibernat- range expansion and thus higher elk densities are likely to ing species, may change less. Coyotes, for instance, are highly occur. We expect, therefore, that the wolf-elk-scavenger dependent on late-winter and early-spring carrion to carry complex will accrue added importance in the years to come. them over until late spring, when elk calves and ground Future studies examining climate change impacts on spring squirrels become abundant. If late-winter carrion were to and summer rainfall, which sets forage levels for elk, will be disappear without a corresponding change in the timing of crucial to further deciphering the effects of global change on elk calving or ground squirrel emergence, a serious food trophic relationships in Yellowstone. bottleneck could develop. We are just beginning to understand the interaction As carrion becomes more concentrated over a shorter between top predators, such as wolves, and global climate window of the year, the relative access to carrion among patterns. On Isle Royale, trophic effects have recently been different scavenger species may change. Highly aggregated or shown to be mediated by behavioral responses to climate. pulsed resources saturate local communities of scavengers, There, gray wolf pack size is partly controlled by climatic allowing species with better recruitment abilities (animals conditions that, in turn, affect wolf kill-rates on moose (Alces capable of covering large distances and communicating about alces) and consequent herbivory levels on balsam fir (Abies the location of resources such as ravens and bald eagles) to balsamea) [30]. In Yellowstone, our scenarios demonstrate that dominate consumption at carcasses [17]. Resources that are wolves act to retard the effects of a changing climate on more dispersed, conversely, do not saturate local scavenger scavenger species. Together these results begin to elucidate communities, so that a competitive dominance hierarchy the expected changes that may occur to boreal ecosystems as (with grizzly bears and coyotes at the top) determines which a result of climate change effects on top predators. PLoS Biology | www.plosbiology.org0574 April 2005 | Volume 3 | Issue 4 | e92 Climate Change Buffers Materials and Methods representing the current distribution of Yellowstone wolves [31]. We then inserted our randomly chosen monthly SDTH values and The northern range of Yellowstone NP is the wintering area of the wolf pack sizes into equation 2 to yield the amount of carrion park’s largest elk herd and home to 4–6 gray wolf packs. Elevations available per month with wolves. For each run of each scenario, we range from 1,500 to 3,400 m, with 87% of the area between 1,500 and recorded the reduction in monthly winter biomass available to 2,400 m [25]. The climate is characterized by short, cool summers and scavengers in 2000 as a proportion of what was available in 1950. long, cold winters, with most annual precipitation falling as snow. Our statistical modeling approach, although rooted empirically, is Mean annual temperature is 1.8 8C, and mean annual precipitation is limited by the fact that it does not take into account the possible 31.7 cm [25]. Large, open valleys of grass meadows and shrub steppe effects of wolf and elk population dynamics on carrion availability. In dominate the landscape, with coniferous forests occurring at higher order to explore these effects, therefore, we used a previously elevations and on north-facing slopes. published model [29] that was originally built to explore the effects of Weather data analysis. Since 1948, meteorological data has been wolf and elk population dynamics on monthly carrion flow to collected daily from two permanent weather stations on the northern scavengers. range of Yellowstone NP. One is located in Mammoth Hot Springs at Wolf effects: Dynamic model. The details of the model are exactly park headquarters near the northern entrance to the park. The other the same as in Wilmers and Getz [29], except for the following is located at the Tower Falls ranger station about 29 km east of changes. In the original model, SDTH was incorporated into the elk Mammoth. Data for the period 01 August 1948 to 01 June 2003 were population dynamics but was treated as a random variable. In the made available to us by the Western Regional Climate Center in present study, we modified the model so that the actual progression Reno, Nevada, United States. of winter weather from 1950 to 2000 was used. We ran the model for Using linear regression, we investigated multiannual trends in 51 y, from 1950 to 2001. We selected SDTH, V, for the year and month monthly average SDTH over the 55 y provided in the data set. SDTH in question from the Tower Falls regression equations in exactly the is treated as the response variable and regressed upon year. We also same manner that we describe above in the statistical model. Since examined trends in the timing of the date of first bare ground. This the distribution of elk among age classes from 1950 is not known, we was defined as the first day of the year for which SDTH was zero. In performed, as a baseline, a 50-y run of the model under average 1950 order to understand changing patterns in SDTH, we analyzed average monthly snowfall (SNFL), average TMIN and TMAX, and the number weather conditions. This is long enough for the effects of initial of days per winter that TMAX exceeded freezing. conditions to dissipate. We then used the numbers and age structure Wolf effects: Statistical model. In order to compare the effects of of the final month of the baseline run as the initial conditions of the carrion availability to scavengers under climate change in scenarios run using observed weather data from 1950 to 2000. with and without wolves, we used previously published regression Sensitivity analyses were conducted using Monte Carlo methods to assess the relative effects of different parameter values on model equations [23] relating SDTH, S, to monthly carrion availability, Cp, prior to wolf reintroduction given by output [29,32]. Since the primary goal of using the dynamic model is to assess whether late-winter carrion will be affected by elk and wolf Cp ¼ 14:48 þ 21:04S ð1Þ population dynamics in the context of a changing climate, we defined an output variable, z, as the percent change in late-winter carrion and relating SDTH and wolf pack size to carrion availability, Ca, after from 1950 to 2000. We assigned March and April to late winter for wolf reintroduction [16] obtained using comparison to Figure 4, since these are the two months showing a ¼ ð ÞðÞ significant effect between scenarios with and without wolves. For each Ca K P 30 1 Q 2 scenario, we conducted 1,000 runs of the model, choosing a different where K is the wolf kill-rate per wolf, P is the wolf pack size, 30 is the set of parameter values at random from the ranges provided in Table number of days in a month, and Q is the percent of the edible biomass 1 of Wilmers and Getz [29]. Each model parameter was then regressed of a carcass consumed by a wolf pack given by Wilmers et al. [16]. We against z to determine its effect. used Monte Carlo methods, as elaborated below, to reconstruct how much carrion would have been available to scavengers during each of the winter months (November through April) in the years 1950 and Acknowledgments 2000 under scenarios with and without wolves. Specifically, for each We thank Dan Stahler, Doug Smith, Blake Suttle, Mary Power, Dale scenario [1950 without wolves, 2000 without wolves, 1950 with wolves, McCullough, and George Wittemeyer for helpful discussions and and 2000 with wolves], we drew 100 random SDTH values for each of the months, where SDTH was assumed to be normally distributed reviews of the manuscript. Reviews by Bob Holt, Andy Dobson, and with mean and standard error for the years 1950 and 2000 given by David Macdonald also greatly improved the manuscript. Funding was the regression analyses of the Tower Falls weather data (see Figure 2). provided by an Environmental Protection Agency Science to Achieve This incorporated uncertainty into our estimate of SDTH for the Results fellowship to CCW and by the James S. McDonnell years 1950 and 2000, allowing us to draw random SDTH values from Foundation Twenty-First Century Science Initiative Study of Com- those years for our Monte Carlo simulation. In the scenarios without plex Systems award to WMG. wolves, we inserted our randomly chosen monthly SDTH values for Competing interests. The authors have declared that no competing each year and each run into equation 1 to yield the amount of carrion interests exist. available per month without wolves. We used the same procedure for Author contributions. CCW conceived and designed the experi- selecting SDTH in our scenario with wolves. In order to select wolf ments, performed the experiments, and analyzed the data. CCW and pack size, we assumed that wolf pack sizes were normally distributed, WMG contributed reagents/materials/analysis tools and wrote the with a mean (6 standard deviation) pack size of 10.6 (6 5) paper. & References 8. Visser ME, Holleman LJM (2001) Warmer springs disrupt the synchrony of 1. Walther G, Post E, Convey P, Menzel A, Parmesan C, et al. (2002) Ecological oak and winter moth phenology. Proc R Soc Lond B Biol Sci 268: 289–294. responses to recent climate change. Nature 416: 389–395. 9. Wilmers CC, Sinha S, Brede M (2002) Examining the effects of species 2. Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, et al., richness on community stability: An assembly model approach. Oikos 99: editors (2001) Climate change 2001: The scientific basis. Cambridge: 363–367. Cambridge University Press. 881 p. 10. McCann K, Hastings A, Huxel GR (1998) Weak trophic interactions and the 3. Barry J, Baxter C, Sagarin R, Gilman S (1995) Climate-related, long-term balance of nature. Nature 395: 794–798. faunal changes in a California rocky intertidal. Science 267: 672–675. 11. Naeem S, Li S (1997) Biodiversity enhances ecosystem reliability. Nature 4. Schmitz OJ, Post E, Burns CE, Johnston KM (2003) Ecosystem response to 390: 507–509. global climate change: Moving beyond color mapping. BioScience 53: 1–7. 12. Tilman D, Wedin D, Knops J (1996) Productivity and sustainability 5. Beebee TJC (1995) Amphibian breeding and climate change. Nature 374: influenced by biodiversity in grassland ecosystems. Nature 379: 718–720. 219–220. 13. Paine RT (1969) A note on trophic complexity and community stability. Am 6. Both C, Visser ME (2001) Adjustment to climate change is constrained by Nat 103: 91–93. arrival date in a long-distance migrant bird. Nature 411: 296–298. 14. Power ME, Tilman D, Estes JA, Menge BA, Bond WJ, et al. (1996) Challenges 7. Visser ME, van Noordwijk AJ, Tinbergen JM, Lessells CM (1998) Warmer in the quest for keystones: Identifying keystone species is difficult—but springs lead to mistimed reproduction in great tits (Parus major). Proc R Soc essential to understanding how loss of species will affect ecosystems. Lond B Biol Sci 265: 1867–1870. BioScience 46: 609–620. PLoS Biology | www.plosbiology.org 0575 April 2005 | Volume 3 | Issue 4 | e92 Climate Change Buffers 15. Bangs EE, Fritts SH (1996) Reintroducing the gray wolf to central Idaho and 24. Parker KL, Robbins CT, Hanley TA (1984) Energy expenditures for Yellowstone National Park. Wildl Soc Bull 24: 402–413. locomotion by mule deer and elk. J Wildl Manage 48: 474–488. 16. Wilmers CC, Crabtree RL, Smith D, Murphy KM, Getz WM (2003) Trophic 25. Houston DB (1982) The northern Yellowstone elk: Ecology and manage- facilitation by introduced top predators: Gray wolf subsidies to scavengers ment. New York: Macmillan. 493 p. in Yellowstone National Park. J Anim Ecol 72: 909–916. 26. Mech LD, Smith DW, Murphy KM, MacNulty DR (2001) Winter severity and 17. Wilmers CC, Stahler DR, Crabtree RL, Smith DW, Getz WM (2003) wolf predation on a formerly wolf-free elk herd. J Wildl Manage 65: 998– Resource dispersion and consumer dominance: Scavenging at wolf- and 1003. hunter-killed carcasses in Greater Yellowstone, USA. Ecol Lett 6: 996–1003. 27. Sagarin R, Micheli F (2001) Climate change in nontraditional data sets. 18. Crabtree RL, Sheldon JW (1999) Coyotes and canid coexistence in Nature 294: 811. Yellowstone. In: Clark TW, Curlee AP, Minta SC, Kareiva PM, editors. 28. Inouye DW, Barr B, Armitage KB, Inouye BD (2000) Climate change is Carnivores in ecosystems: The Yellowstone experience. New Haven: Yale affecting altitudinal migrants and hibernating species. Proc R Soc Lond B Biol Sci 97: 1630–1633. University Press. pp. 429. 29. Wilmers CC, Getz WM (2004) Simulating the effects of wolf-elk population 19. Blanchard BM (1987) Size and growth patterns of the Yellowstone grizzly dynamics on resource flow to scavengers. Ecol Modell 177: 193–208. bear. International Conference of Bear Research and Management 7: 99– 30. Post E, Peterson RO, Stenseth NC, McLaren BE (1999) Ecosystem 107. consequences of wolf behavioural response to climate. Nature 401: 905– 20. Newton I, Davis PE, Davis JE (1982) Ravens and buzzards in relation to 907. sheep-farming and forestry in Wales. J Appl Ecol 19: 681–706. 31. Smith DW, Stahler DR, Guernsey DS (2003) Yellowstone wolf project: 21. Swenson JE, Alt KL, Eng RL (1986) Ecology of bald eagles in the Greater Annual report 2002. National Park Service, Yellowstone Center for Yellowstone ecosystem. Wildl Monogr (95): 1–46. Resources, Yellowstone National Park, Wyoming. Report Number YCR- 22. Dhindsa MS, Boag DA (1990) The effect of food supplementation on the NR-2003–04. 26 p. Available: http://www.nps.gov/yell/nature/animals/wolf/ reproductive success of black-billed magpies Pica pica. Ibis 132: 595–602. wolfup.html. 23. Gese EM, Ruff RL, Crabtree RL (1996) Foraging ecology of coyotes (Canis 32. Wisdom MJ, Mills LS, Doak DF (2000) Life-stage simulation analysis: latrans): The influence of extrinsic factors and a dominance hierarchy. Can J Estimating vital-rate effects on population growth for conservation. Zool 74: 769–783. Ecology 81: 628–641. PLoS Biology | www.plosbiology.org 0576 April 2005 | Volume 3 | Issue 4 | e92